Citation
Measuring Demand Factors Influencing Market Penetration and Buying Frequency for Flowers

Material Information

Title:
Measuring Demand Factors Influencing Market Penetration and Buying Frequency for Flowers
Creator:
GARCIA, MARCO ANTONION PALMA
Copyright Date:
2008

Subjects

Subjects / Keywords:
Age groups ( jstor )
Arithmetic mean ( jstor )
Consumer spending ( jstor )
Floral design ( jstor )
Flowers ( jstor )
Market penetration ( jstor )
Mathematical variables ( jstor )
Modeling ( jstor )
Parametric models ( jstor )
Prices ( jstor )
City of Gainesville ( local )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Marco Antonion Palma Garcia. Permission granted to University of Florida to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
4/17/2006
Resource Identifier:
76700341 ( OCLC )

Downloads

This item has the following downloads:

palma_m ( .pdf )

palma_m_Page_044.txt

palma_m_Page_068.txt

palma_m_Page_197.txt

palma_m_Page_177.txt

palma_m_Page_163.txt

palma_m_Page_182.txt

palma_m_Page_113.txt

palma_m_Page_203.txt

palma_m_Page_104.txt

palma_m_Page_256.txt

palma_m_Page_235.txt

palma_m_Page_252.txt

palma_m_Page_272.txt

palma_m_Page_278.txt

palma_m_Page_212.txt

palma_m_Page_191.txt

palma_m_Page_232.txt

palma_m_Page_121.txt

palma_m_Page_247.txt

palma_m_Page_077.txt

palma_m_Page_217.txt

palma_m_Page_181.txt

palma_m_Page_164.txt

palma_m_Page_056.txt

palma_m_Page_211.txt

palma_m_Page_083.txt

palma_m_Page_043.txt

palma_m_Page_071.txt

palma_m_Page_172.txt

palma_m_Page_205.txt

palma_m_Page_130.txt

palma_m_Page_133.txt

palma_m_Page_266.txt

palma_m_Page_208.txt

palma_m_Page_015.txt

palma_m_Page_084.txt

palma_m_Page_193.txt

palma_m_Page_048.txt

palma_m_Page_138.txt

palma_m_Page_117.txt

palma_m_Page_259.txt

palma_m_Page_175.txt

palma_m_Page_105.txt

palma_m_Page_099.txt

palma_m_Page_057.txt

palma_m_Page_204.txt

palma_m_Page_050.txt

palma_m_Page_075.txt

palma_m_Page_142.txt

palma_m_Page_025.txt

palma_m_Page_226.txt

palma_m_Page_132.txt

palma_m_Page_224.txt

palma_m_Page_244.txt

palma_m_Page_137.txt

palma_m_Page_151.txt

palma_m_Page_166.txt

palma_m_Page_127.txt

palma_m_Page_192.txt

palma_m_Page_016.txt

palma_m_Page_028.txt

palma_m_Page_023.txt

palma_m_Page_140.txt

palma_m_Page_237.txt

palma_m_Page_035.txt

palma_m_Page_260.txt

palma_m_Page_022.txt

palma_m_Page_045.txt

palma_m_Page_267.txt

palma_m_Page_129.txt

palma_m_Page_128.txt

palma_m_Page_002.txt

palma_m_Page_089.txt

palma_m_Page_139.txt

palma_m_Page_276.txt

palma_m_Page_008.txt

palma_m_Page_169.txt

palma_m_Page_011.txt

palma_m_Page_176.txt

palma_m_Page_141.txt

palma_m_Page_041.txt

palma_m_Page_198.txt

palma_m_Page_085.txt

palma_m_Page_069.txt

palma_m_Page_026.txt

palma_m_Page_020.txt

palma_m_Page_183.txt

palma_m_Page_052.txt

palma_m_Page_033.txt

palma_m_Page_120.txt

palma_m_Page_190.txt

palma_m_Page_029.txt

palma_m_Page_161.txt

palma_m_Page_062.txt

palma_m_Page_145.txt

palma_m_Page_144.txt

palma_m_Page_080.txt

palma_m_Page_230.txt

palma_m_Page_037.txt

palma_m_Page_030.txt

palma_m_Page_222.txt

palma_m_Page_059.txt

palma_m_Page_277.txt

palma_m_Page_098.txt

palma_m_Page_250.txt

palma_m_Page_021.txt

palma_m_Page_215.txt

palma_m_Page_157.txt

palma_m_Page_257.txt

palma_m_Page_273.txt

palma_m_Page_060.txt

palma_m_Page_227.txt

palma_m_Page_147.txt

palma_m_Page_171.txt

palma_m_Page_261.txt

palma_m_Page_210.txt

palma_m_Page_017.txt

palma_m_Page_018.txt

palma_m_Page_146.txt

palma_m_Page_066.txt

palma_m_Page_202.txt

palma_m_Page_152.txt

palma_m_Page_032.txt

palma_m_Page_243.txt

palma_m_Page_236.txt

palma_m_Page_081.txt

palma_m_Page_012.txt

palma_m_Page_095.txt

palma_m_Page_219.txt

palma_m_Page_156.txt

palma_m_Page_067.txt

palma_m_Page_076.txt

palma_m_Page_233.txt

palma_m_Page_240.txt

palma_m_Page_086.txt

palma_m_Page_074.txt

palma_m_Page_135.txt

palma_m_Page_009.txt

palma_m_Page_238.txt

palma_m_Page_218.txt

palma_m_Page_201.txt

palma_m_Page_082.txt

palma_m_Page_185.txt

palma_m_Page_271.txt

palma_m_Page_078.txt

palma_m_Page_063.txt

palma_m_Page_234.txt

palma_m_Page_014.txt

palma_m_Page_246.txt

palma_m_Page_264.txt

palma_m_Page_091.txt

palma_m_Page_188.txt

palma_m_Page_126.txt

palma_m_Page_254.txt

palma_m_Page_087.txt

palma_m_Page_053.txt

palma_m_Page_072.txt

palma_m_Page_134.txt

palma_m_Page_097.txt

palma_m_Page_143.txt

palma_m_Page_027.txt

palma_m_Page_150.txt

palma_m_Page_207.txt

palma_m_Page_206.txt

palma_m_Page_136.txt

palma_m_Page_199.txt

palma_m_Page_262.txt

palma_m_Page_231.txt

palma_m_Page_279.txt

palma_m_Page_107.txt

palma_m_Page_003.txt

palma_m_Page_123.txt

palma_m_Page_195.txt

palma_m_Page_258.txt

palma_m_Page_019.txt

palma_m_Page_148.txt

palma_m_Page_102.txt

palma_m_Page_109.txt

palma_m_Page_228.txt

palma_m_Page_162.txt

palma_m_Page_073.txt

palma_m_Page_223.txt

palma_m_Page_031.txt

palma_m_Page_270.txt

palma_m_Page_122.txt

palma_m_Page_090.txt

palma_m_Page_155.txt

palma_m_Page_006.txt

palma_m_Page_064.txt

palma_m_Page_200.txt

palma_m_Page_046.txt

palma_m_Page_096.txt

palma_m_Page_149.txt

palma_m_Page_042.txt

palma_m_Page_036.txt

palma_m_Page_154.txt

palma_m_Page_010.txt

palma_m_Page_184.txt

palma_m_Page_221.txt

palma_m_Page_269.txt

palma_m_Page_119.txt

palma_m_Page_112.txt

palma_m_Page_242.txt

palma_m_Page_106.txt

palma_m_Page_001.txt

palma_m_Page_265.txt

palma_m_Page_196.txt

palma_m_Page_160.txt

palma_m_Page_110.txt

palma_m_Page_040.txt

palma_m_Page_274.txt

palma_m_Page_108.txt

palma_m_Page_275.txt

palma_m_Page_268.txt

palma_m_Page_220.txt

palma_m_Page_024.txt

palma_m_Page_186.txt

palma_m_Page_245.txt

palma_m_pdf.txt

palma_m_Page_225.txt

palma_m_Page_013.txt

palma_m_Page_159.txt

palma_m_Page_170.txt

palma_m_Page_174.txt

palma_m_Page_214.txt

palma_m_Page_005.txt

palma_m_Page_241.txt

palma_m_Page_173.txt

palma_m_Page_038.txt

palma_m_Page_239.txt

palma_m_Page_158.txt

palma_m_Page_125.txt

palma_m_Page_179.txt

palma_m_Page_103.txt

palma_m_Page_209.txt

palma_m_Page_088.txt

palma_m_Page_251.txt

palma_m_Page_187.txt

palma_m_Page_153.txt

palma_m_Page_255.txt

palma_m_Page_248.txt

palma_m_Page_004.txt

palma_m_Page_061.txt

palma_m_Page_178.txt

palma_m_Page_047.txt

palma_m_Page_115.txt

palma_m_Page_131.txt

palma_m_Page_094.txt

palma_m_Page_118.txt

palma_m_Page_229.txt

palma_m_Page_216.txt

palma_m_Page_111.txt

palma_m_Page_055.txt

palma_m_Page_100.txt

palma_m_Page_054.txt

palma_m_Page_079.txt

palma_m_Page_189.txt

palma_m_Page_070.txt

palma_m_Page_007.txt

palma_m_Page_253.txt

palma_m_Page_065.txt

palma_m_Page_167.txt

palma_m_Page_180.txt

palma_m_Page_249.txt

palma_m_Page_051.txt

palma_m_Page_194.txt

palma_m_Page_058.txt

palma_m_Page_093.txt

palma_m_Page_263.txt

palma_m_Page_116.txt

palma_m_Page_049.txt

palma_m_Page_101.txt

palma_m_Page_114.txt

palma_m_Page_039.txt

palma_m_Page_034.txt

palma_m_Page_168.txt

palma_m_Page_213.txt

palma_m_Page_165.txt

palma_m_Page_124.txt

palma_m_Page_092.txt


Full Text












MEASURING DEMAND FACTORS INFLUENCING MARKET PENETRATION
AND BUYING FREQUENCY FOR FLOWERS















By

MARCO ANTONIO PALMA GARCIA


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005

































Copyright 2005

by

Marco Antonio Palma Garcia

































This document is dedicated to my parents Antonia and Marco, and my siblings Eric and
Vanessa, with gratitude and love.















ACKNOWLEDGMENTS

I will begin by thanking the Almighty for giving me life, and this educational

opportunity. This work would not have been possible without the continuous support,

guidance, patience and time dedicated from my committee chair, Dr. Ronald W. Ward. I

also want to extend my most sincere appreciation to the rest of my committee members

for their support: Dr. Lisa House, Dr. Mark Brown, Dr. Tom Sheehan and Dr. Gerhard

Schiefer.

I would like to thank my family, specially Antonia, Marco, Vanessa, Roy, Carolina,

Eric Sebastian, Ivette, Alicia, Doris and Monica, for their affection, love, encouragement

and support always, specially during difficult times. A special thank you goes to my

friends Fausto, Ricky, Jorge, Angel, Sergio, Claudio, and my friends in Gainesville, for

being a source of strength and motivation during my graduate school experience.

Finally I would like to thank my classmates Shiferaw, Mariana, Angel, Lurleen,

Ronald, Joy, and Raphael for their continuous friendship, and support during all of my

graduate school experience, especially during the core of the program. I also want to

thank the faculty and staff of the Food and Resource Economics Department of the

University of Florida for their assistance, especially Dr. Charles Adams, Dr. Richard

Beilock, Dr. Tom Spreen, Dr. William Messina, Dr. Jane Luzar, Dr. JeffBurkhardt, Brian

Sevier and Jessica Herman.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ................................................................................................. iv

LIST OF TABLES .................................................... ........ .. .............. viii

LIST OF FIGURES ......... ....................... .......... ....... ............ xiv

ABSTRACT .............. .......................................... xix

CHAPTER


1 IN TR OD U CTION ............................................... .. ......................... ..

O overview of the Industry ......................................................... ............................ 3
Problem Statem ent .................. ................................ .. .... ...... ....... ...... .. 10
O bj ectives .................................... .................. ................... .............. 11
R research M methodology ...................................... ............. ...... ........................ .. 11
D ata an d S co p e ...................................................... ................ 13
O organization of the Study ................................................... ........ ............... .14

2 LITERATURE REVIEW ........................................................................... 15

Consum er Behavior .................. ............................ ... ..... ................. 15
Consumer Demand Analysis ................................................. 20
Utility Maximization and Demand Functions ............................................... 20
Properties of the D em and Functions ................................................................ 26
A dding-up restriction ............................................................................26
Sym m etry restriction ........................................................ ............. 27
N egativity restriction ......................................................... ............... 27
H om ogeneity restriction................................................... ................. 28
M marketing Research M odels ......................................................... .............. 28
M market Penetration M odels ........................................ ........................... 29
L ogistic function .................................. ................. ..... ....... 29
The Pyatt function ............................ ..... ...... .... ............... .30
The G om pertz function ........................................ .......................... 31
T he W eblus m odel ........................... ................ .............. ...............




v









The log-inverse function ........................................ ......................... 32
The lognorm al function ........... .......................................... ...... ......... 33
R epeat B buying M models ............................................................. ............... 33
Review of Past Studies on Flower Products.......................................................36

3 UNITED STATES FLORAL INDUSTRY .....................................................43

In tro d u ctio n ........................................................................................................... 4 3
Expenditures on Flow ers .................................................. .............................. 45
Transactions on Flow ers ............................................................ ............ 50
Expenditures per Transaction ............................................................................. 56
E expenditures P er B uyer ................................................................... .....................60
M market Penetration ............................................... ...................... 64

4 CONCEPTUAL FRAMEWORK AND THEORETICAL MODELS ......................69

Consumer Demand Theory for the Case of Flowers .............................. ...............69
An Overview of NPD D ata Set...................................................... ...................71
T h e T o b it M o d el ................................................................................................... 7 5
M market P enetration M odels....................................... ............................................78
Penetration M odel I ....................................... ....... ........... .... ... 79
Penetration Model II..................................... .............. 79
B uyer F requency M odels............................................................ ..........................80
Frequency M odel I ...................................... ........... ......... ... ...8 81
Frequency M odel II ........................................... .. ..... .... .......... .... 82

5 EM PIRICAL RE SU LTS .................................................. .............................. 84

D ata U sa g e ........................................................................................8 4
D em and M odel Equations ................................................ .............................. 85
Model I Results ...................................................... ......................87
M market Penetration M odel Results I....................................... ............... 87
Buyer Frequency M odel R results I................................................................. 93
M odel II Results ............................................................ ......... 99
M market Penetration M odel Results II ..................................... ............... 100
Buyer Frequency Model Results II....................................................... 109

6 SIM U LA TION A N A LY SIS ............................................................... ............... 19

Introdu action ................................................................................................ ..... 119
Simulations For Cut-Flowers.................................... ............123
Simulations For Flowering Plants And Greens ....................................................... 131
Simulations For Dry/Artificial Flowers ......................................... ..............139
Sim ulations F or O outdoor ................................................. ................................... 147

7 SUMMARY AND CONCLUSIONS ..................................... ..........157









Sum m ary and Conclusions ............... ......................................... .............. 157
Limitations and Direction for Future Research ....................................................... 167

APPENDIX

A M O D E L I R E SU L T S ....................................................................... ..................169

B T SP P R O G R A M S ........................................................................... ...................24 1

LIST OF REFEREN CES ............................................................ .. ............... 256

B IO G R A PH IC A L SK E T C H ........................................ ............................................259
















LIST OF TABLES


Table pge

4.1 Percentage of Observations of Penetration Model Dependent Variable That
Are Censored at Zero. Source: AFE and Ipsos Group. .........................................74

4.2 Percentage of Observations of Frequency Model Dependent Variable That Are
Censored at One. Source: AFE and Ipsos Group. .................................................75

4.3 Variables for the M market Penetration M odel I ............................... ............... .80

4.4 Variables for the M market Penetration M odel II................................................ 81

5.1 Distribution of the States for Each Region.................................... ..........86

5.2 General Statistical Information About the Market Penetration Model by Flower
T y p e ...................................... ..................................... ................ 8 8

5.3 Market Penetration Parameter Estimates and T-Values for Indoor and Cut-
F low ers. ..............................................................................89

5.4 Market Penetration Parameter Estimates and T-Values for Flower
Arrangements and Non-Arrangements................. ...............................................90

5.5 Market Penetration Parameter Estimates and T-Values for Plants and
D ry/A artificial Flow ers. ...... ........................... .......................................... 91

5.6 Market Penetration Parameter Estimates and T-Values for Outdoor Flowers.........92

5.7 General Statistical Information About the Buyer Frequency Model by Flower
T y p e ......... ................... ......... ......................................................9 5

5.8 Buyer Frequency Parameter Estimates and T-values for Indoor and Cut-
F low ers. ..............................................................................96

5.9 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements
and N on-A rrangem ents. ................................................ ............................... 96

5.10 Buyer Frequency Parameter Estimates and T-values for Plants and
D ry/A artificial ......... .... .............................. .......... ............. 97

5.11 Buyer Frequency Parameter Estimates and T-values for Outdoor...........................97









5.12 General Statistical Information About the Market Penetration Model II by
F low er T ype. ...................................................................... 100

5.13 Market Penetration Parameter Estimates and T-Values for Indoor and Cut-
F low ers. .............................................................................10 1

5.14 Market Penetration Parameter Estimates and T-Values for Flower
Arrangements and Non-Arrangements....................................... ............... 102

5.15 Market Penetration Parameter Estimates and T-Values for Plants and
Dry/Artificial Flow ers. ...... ........................... ........................................103

5.16 Market Penetration Parameter Estimates and T-Values for Outdoor Flowers.......104

5.17 General Statistical Information About the Buyer Frequency Model II by
F low er T ype. ...................................................................... 109

5.18 Buyer Frequency Parameter Estimates and T-values for Indoor and Cut-
F low ers. ................................................................................ ......... 110

5.19 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements
and N on-A rrangem ents. .................................................................................... 111

5.20 Buyer Frequency Parameter Estimates and T-values for Plants and
D ry/A artificial ........... .. ................. ....... ........ ....... ........ ....... 112

5.21 Buyer Frequency Parameter Estimates and T-values for Outdoor.........................113

A. 1 Market Penetration Model I Results for Indoor and Cut-Flowers in New
E nglan d .......................................................... ........................... 169

A.2 Market Penetration Model I Results for Flower Arrangements and Non-
A rrangem ents in N ew England ........................................ ......................... 170

A.3 Market Penetration Model I Results for Plants and Dry/Artificial in New
E n g lan d .......................... .. ......... ... .. .......... .......................................17 1

A.4 Market Penetration Model I Results for Outdoor in New England.......................172

A.5 Buyer Frequency Model I Results for Indoor and Cut-Flowers in New England.. 173

A.6 Buyer Frequency Model I Results for Flower Arrangements and Non-
A rrangem ents in N ew England ........................................ ......................... 174

A.7 Buyer Frequency Model I Results for Plants and Dry/Artificial in New
E ngland ............... ........... .......................... ............................175

A.8 Buyer Frequency Model I Results for Outdoor in New England.........................176









A.9 Market Penetration Model I Results for Indoor and Cut-Flowers in Middle
A tlantic ............... ......... ............................................. ................ ........ 177

A. 10 Market Penetration Model I Results for Flower Arrangements and Non-
Arrangem ents in M iddle Atlantic...................................... ......................... 178

A. 11 Market Penetration Model I Results for Plants and Dry/Artificial in Middle
A tlantic .......................................... ............. ............................ 179

A. 12 Market Penetration Model I Results for Outdoor in Middle Atlantic ....................180

A. 13 Buyer Frequency Model I Results for Indoor and Cut-Flowers in Middle
A tlantic ................................................................ ..... ..... ......... 181

A. 14 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangem ents in M iddle Atlantic...................................... ......................... 182

A. 15 Buyer Frequency Model I Results for Plants and Dry/Artificial in Middle
A tlan tic ...................... .. .. ......... .. .. .................................................18 3

A. 16 Buyer Frequency Model I Results for Outdoor in Middle Atlantic .....................184

A. 17 Market Penetration Model I Results for Indoor and Cut-Flowers in East North
C e n tra l .......................................................................... 1 8 5

A. 18 Market Penetration Model I Results for Flower Arrangements and Non-
Arrangem ents in East N orth Central ........................................... ............... 186

A. 19 Market Penetration Model I Results for Plants and Dry/Artificial in East North
C e n tra l .......................................................................... 1 8 7

A.20 Market Penetration Model I Results for Outdoor in East North Central .............188

A.21 Buyer Frequency Model I Results for Indoor and Cut-Flowers in East North
C e n tra l .......................................................................... 1 8 9

A.22 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangem ents in East N orth Central ........................................... ............... 190

A.23 Buyer Frequency Model I Results for Plants and Dry/Artificial in East North
C e n tra l ............................................................................ 1 9 1

A.24 Buyer Frequency Model I Results for Outdoor in East North Central................. 192

A.25 Market Penetration Model I Results for Indoor and Cut-Flowers in West North
C e n tra l .......................................................................... 1 9 3

A.26 Market Penetration Model I Results for Flower Arrangements and Non-
Arrangements in West North Central..... .................... ...............194









A.27 Market Penetration Model I Results for Plants and Dry/Artificial in West North
C e n tra l .......................................................................... 1 9 5

A.28 Market Penetration Model I Results for Outdoor in West North Central ............196

A.29 Buyer Frequency Model I Results for Indoor and Cut-Flowers in West North
C e n tra l .......................................................................... 1 9 7

A.30 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangements in West North Central ...........................................................198

A.31 Buyer Frequency Model I Results for Plants and Dry/Artificial in West North
C e n tra l .......................................................................... 1 9 9

A.32 Buyer Frequency Model I Results for Outdoor in West North Central ...............200

A.33 Market Penetration Model I Results for Indoor and Cut-Flowers in South
A tlantic ............................................................... ..... ..... ......... 201

A.34 Market Penetration Model I Results for Flower Arrangements and Non-
A rrangem ents in South A tlantic................................... ............................. ....... 202

A.35 Market Penetration Model I Results for Plants and Dry/Artificial in South
A tlantic ............................................................... ..... ..... ......... 203

A.36 Market Penetration Model I Results for Outdoor in South Atlantic ....................204

A.37 Buyer Frequency Model I Results for Indoor and Cut-Flowers in South
A tlantic ............................................................... ..... ..... ......... 205

A.38 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangem ents in South Atlantic................................................................. ...... 206

A.39 Buyer Frequency Model I Results for Plants and Dry/Artificial in South
A tlantic ............................................................... ..... ..... ......... 207

A.40 Buyer Frequency Model I Results for Outdoor in South Atlantic .......................208

A.41 Market Penetration Model I Results for Indoor and Cut-Flowers in East South
C e n tra l .......................................................................... 2 0 9

A.42 Market Penetration Model I Results for Flower Arrangements and Non-
Arrangem ents in East South Central ........................................... ............... 210

A.43 Market Penetration Model I Results for Plants and Dry/Artificial in East South
Central ..................................... ................................... ......... 211

A.44 Market Penetration Model I Results for Outdoor in East South Central .............212









A.45 Buyer Frequency Model I Results for Indoor and Cut-Flowers in East South
C e n tra l .......................................................................... 2 1 3

A.46 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangem ents in East South Central ........................................... ............... 214

A.47 Buyer Frequency Model I Results for Plants and Dry/Artificial in East South
C e n tra l .......................................................................... 2 1 5

A.48 Buyer Frequency Model I Results for Outdoor in East South Central...................216

A.49 Market Penetration Model I Results for Indoor and Cut-Flowers in West South
C e n tra l .......................................................................... 2 1 7

A.50 Market Penetration Model I Results for Flower Arrangements and Non-
Arrangements in West South Central ...........................................................218

A.51 Market Penetration Model I Results for Plants and Dry/Artificial in West South
C e n tra l .......................................................................... 2 1 9

A.52 Market Penetration Model I Results for Outdoor in West South Central .............220

A.53 Buyer Frequency Model I Results for Indoor and Cut-Flowers in West South
C e n tra l .......................................................................... 2 2 1

A.54 Buyer Frequency Model I Results for Flower Arrangements and Non-
Arrangements in West South Central ...........................................................222

A.55 Buyer Frequency Model I Results for Plants and Dry/Artificial in West South
C e n tra l .......................................................................... 2 2 3

A.56 Buyer Frequency Model I Results for Outdoor in West South Central .................224

A.57 Market Penetration Model I Results for Indoor and Cut-Flowers in Mountain.....225

A.58 Market Penetration Model I Results for Flower Arrangements and Non-
A rrangem ents in M ountain.......................................................... ............... 226

A.59 Market Penetration Model I Results for Plants and Dry/Artificial in Mountain....227

A.60 Market Penetration Model I Results for Outdoor in Mountain..............................228

A.61 Buyer Frequency Model I Results for Indoor and Cut-Flowers in Mountain........229

A.62 Buyer Frequency Model I Results for Flower Arrangements and Non-
A rrangem ents in M ountain.......................................................... ............... 230

A.63 Buyer Frequency Model I Results for Plants and Dry/Artificial in Mountain.......231









A.64 Buyer Frequency Model I Results for Outdoor in Mountain..............................232

A.65 Market Penetration Model I Results for Indoor and Cut-Flowers in Pacific .........233

A.66 Market Penetration Model I Results for Flower Arrangements and Non-
A rrangem ents in Pacific .......................................................... ............... 234

A.67 Market Penetration Model I Results for Plants and Dry/Artificial in Pacific ........235

A.68 Market Penetration Model I Results for Outdoor in Pacific ...............................236

A.69 Buyer Frequency Model I Results for Indoor and Cut-Flowers in Pacific ............237

A.70 Buyer Frequency Model I Results for Flower Arrangements and Non-
A rrangem ents in Pacific .......................................................... ............... 238

A.71 Buyer Frequency Model I Results for Plants and Dry/Artificial in Pacific .........239

A.72 Buyer Frequency Model I Results for Outdoor in Pacific ...................................240
















LIST OF FIGURES


Figure page

1.1 Large Grower Sales by Crop Group From 1985 to 2003. Source. Economic
Research Service, U SDA, 2003. ........................... ................. .... ...........5

1.2 Total Number of Large Growers in the US From 1992 to 2003. Source.
Economic Research Service, USDA, 2003 ..................... ....... ............

1.3 Average Sales Per Large Grower by Crop Group from 1992 to 2003. Source.
Economic Research Service, USDA, 2003 ..................... ....... ............

1.4 US Total Imports and Exports from 1976 to 2003. Source. Economic Research
Service, USDA, 2003 ........... .......... .......... .. ............ ....... ....... 6

1.5 US Imports of Cut Flowers and Nursery Crops by Country, 2003. Source.
Economic Research Service, USDA, 2003 ..................... ....... ............

1.6 US Exports of Cut Flowers and Nursery Crops by Country, 2003. Source.
Economic Research Service, USDA, 2003 ..................... ....... ............

2.1 Graphic Presentation of Indifference Curves................. ................... ...... ......... 18

3.1 Total Number of Households by Region from 1993 to 2003. Source: AFE and
Ip sos-N P D group .......................... ...................................... ...... ... .... ... 44

3.2 Total Household Expenditures on Flowers by Region from 1993 to 2003.
Source: AFE and Ipsos-NPD group. ............................................. ............... 45

3.3 Total Household Expenditures on Flowers by Flower Type, from 1993 to
2003. Source: AFE and Ipsos-NPD group. ................ ............ ........ ........... 46

3.4 Total Household Expenditures on Flowers by flower type, Share of Cut-
Flowers Expenditures from 1993 to 2003. Source: AFE and Ipsos-NPD group. ....46

3.5 Total Household Expenditures on Flowers by Purpose from 1993 to 2003.
Source AFE and Ipsos-NPD group. .............................................. ............... 47

3.6 Total Household Expenditures on Flowers by Gender from 1993 to 2003.
Source: AFE and Ipsos-NPD group. ............................................. ............... 48









3.7 Total Household Expenditures on Flowers by Age Groups from 1993 to 2003.
Source: AFE and Ipsos-NPD group. ............................................. ............... 48

3.8 Total Household Expenditures on Flowers by Income Groups from 1993 to
2003. Source: AFE and Ipsos-NPD group. ............................................................49

3.9 Shares of Seasonal Expenditures on Flowers by Flower Type during the 1993
to 2003 Period. Source: AFE and Ipsos-NPD group..............................................50

3.10 Number of Transactions by Month and by Flower Type During the 1993 to
2003 Period. Source: AFE and Ipsos-NPD group ...............................................51

3.11 Number of Transactions by Purpose During the 1993 to 2003 period. Source:
AFE and Ipsos-N PD group. ............................................ ............................. 52

3.12 Number of Transactions by Gender During the 1993 to 2003 period. Source:
AFE and Ipsos-N PD group. ............................................ ............................. 52

3.13 Number of transactions by Age Groups for the 1993 to 2003 Period. Source:
A FE and Ipsos-N PD G roup........................................................... ............... 53

3.14 Number of Transactions by Income Groups for the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 54

3.15 Seasonality of the Number of Transactions per Month for the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 55

3.16 Number of Transactions by Region from the 1993 to 2003 Period. Source:
A FE and Ipsos-N PD G roup........................................................... ............... 55

3.17 Expenditures per transaction by Flower Type During the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 56

3.18 Expenditures per Transaction by Purpose During the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 57

3.19 Expenditures per Transaction by Gender During the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 57

3.20 Average Expenditures per Transaction by Age Groups During the 1993 to
2003 Period. Source: AFE and Ipsos-NPD Group .............................................58

3.21 Average Expenditures per Transaction by Income Groups During the 1993 to
2003 Period. Source: AFE and Ipsos-NPD Group .............................................59

3.22 Average Expenditure per Transaction by Region during the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 59









3.23 Average Expenditures by Buyers by Flower Type During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 60

3.24 Average Expenditures by Buyers by Purpose During the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 61

3.25 Average Expenditures by Buyers by Gender During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 61

3.26 Average Expenditures by Buyers by Age Group During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 62

3.27 Average Expenditures by Buyers by Income Groups During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 63

3.28 Average Expenditures by Buyers by Region During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 64

3.29 Percent of Market Penetration by Flower Type During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 65

3.30 Percent of Market Penetration by Purpose During 1993 to 2003 Period. Source:
A FE and Ipsos-N PD G roup........................................................... ............... 65

3.31 Percentage Market Penetration by Gender During 1993 to 2003 Period. Source:
A FE and Ipsos-N PD G roup........................................................... ............... 66

3.32 Percentage of Market Penetration by Age Groups During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group. ............................................ ............... 67

3.33 Percentage of Market Penetration by Income Groups During 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 67

3.34 Percentage of Market Penetration Index by Region During 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group. .................... .......................... 68

5.1 Flower Type Groups...................... ..............................85

5.2 Market Penetration Seasonality for Cut-Flowers and Plants. .................................93

5.3 Market Penetration Seasonality for Dry/Artificial and Outdoor flowers ...............94

5.4 Buyer Frequency Seasonality for Cut-Flowers and Plants....................................98

5.5 Buyer Frequency Seasonality for Dry/Artificial and Outdoor Flowers .................99

5.6 Market Penetration Seasonality for Cut-Flowers and Plants. .............................105

5.7 Market Penetration Seasonality for Dry/Artificial and Outdoor flowers.............106









5.8 Market Penetration Regional Changes for Cut-Flowers and Plants.......................107

5.9 Market Penetration Regional Changes for Dry/Artificial and Outdoor Flowers. ..108

5.10 Buyer Frequency Seasonality for Cut-Flowers and Plants...................................114

5.11 Buyer Frequency Seasonality for Dry/Artificial and Outdoor Flowers ...............115

5.12 Buyer Frequency Regional Changes for Cut-Flowers and Plants..........................116

5.13 Buyer Frequency Regional Changes for Dry/Artificial and Outdoor Flowers. .....117

6.1 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Age...........................................123

6.2 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Gender...................... ...........124

6.3 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Purpose................................... 126

6.4 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Income ......................................127

6.5 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Income...................... ...........128

6.6 Cut-Flowers Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Income...................... ...........129

6.7 Ranges And Percentages of Variable Changes Affecting Transactions Due to
Frequency of Buying for Cut-Flowers. ..................................... ............... 130

6.8 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Age. ...................................... ............... 131

6.9 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Gender. .................................... ..................... 132

6.10 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their Means for Purpose. ........................... ................................ 134

6.11 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Income. ................................... .................135

6.12 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Income. ........................... ............................ 136









6.13 Plants Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their Means for Income. .......................................................137

6.14 Ranges And Percentages of Variable Changes Affecting the Number of
Transactions Due to Frequency of Buying for Plants ........................................138

6.15 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Age...................................139

6.16 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Gender................ ................ 140

6.17 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Purpose.............................. 142

6.18 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Income ......................................143

6.19 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Seasonality.............................144

6.20 Dry/Artificial Market Penetration, Buyer Frequency and Number of
Transactions Deviations From Their Means for Regions .....................................145

6.21 Ranges And Percentages of Variable Changes Affecting the Number of
Transactions Due to Frequency of Buying for Dry ............ .................146

6.22 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Age. ...................................... ............... 147

6.23 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Gender. .................................... ..................... 148

6.24 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their Means for Purpose. ........................... ................................ 150

6.25 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their M eans for Income. ................................... .................151

6.26 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their Means for Income. .......................................................152

6.27 Outdoor Market Penetration, Buyer Frequency and Number of Transactions
Deviations From Their Means for Income. .......................................................153

6.28 Ranges And Percentages of Variable Changes Affecting the Number of
Transactions Due to Frequency of Buying for Outdoor ............... ...............154

6.29 Percentage of Transactions Due to Frequency of Buying For All Flower Types. .155


xviii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

MEASURING DEMAND FACTORS INFLUENCING MARKET PENETRATION
AND BUYING FREQUENCY FOR FLOWERS

By

Marco Antonio Palma Garcia

December, 2005

Chair: Ronald W. Ward
Major Department: Food and Resource Economics

The floriculture industry is one of the fastest growing sectors of agriculture in the

United States. In 1996, floriculture ranked seventh among commodity groups, behind

only cattle and calves, dairy products, corn, hogs, and soybeans. Floriculture crops,

defined as cut flowers, cut cultivated greens, potted flowering plants, potted foliage,

bedding and garden plants, accounted for about one third of grower cash receipts for

floriculture and environmental horticulture. In order to continue this growing trend it is of

vital importance to gain insight into consumer preferences on floral products.

Specifically, there are two key factors that need to be analyzed in order to understand

how consumers base their decision to buy or not to buy floral products: market

penetration and buyer frequency. Understanding what are the factors that influence non-

buyers of floral products to become buyers, and the factors that influence buyers to

increase their expenditures on floral products is vital information that the industry can use









to design specific programs targeting different demographic groups according to their

specific preferences on flowers.














CHAPTER 1
INTRODUCTION

Consumption behavior has always been of great importance and a topic of focus for

researchers. This importance may be attributed to the relatively strong theoretical basis

for the various consumption hypotheses and an interest in empirical tests of the

underlying propositions. Analysis of consumer demand has always played an important

role in economic theory; this fact is evidenced by the extensive literature that exists on

demand and utility (Johnson et al., 1984).

The consumption of goods takes place because of the satisfaction that the goods or

services provide. The consumption of traditional agricultural food products depends on

the characteristics of the product or attributes that can be measured or quantified. For

example, milk can be measured in the quantity of calories or fat percentage. In contrast to

food products, many nonfood products are consumed because of their aesthetic value.

Flowers are purchased for various reasons such as expression of love or friendship, a way

to express thankfulness or appreciation, beautification purposes for self, or gifts. The

attributes of flowers, or more generally nonfood products, cannot be quantified; therefore

the satisfaction gained from the consumption of these goods is closely related to the

objective of the purchase. This situation also implies that the demand for these products

can be influenced by the characteristics or preferences of the buyers and the reason for

buying the products. This fact can be viewed during special calendar occasions (i.e.,

Mother's Day, Valentine's, etc) where the consumption of floral products is substantially

higher compared to non-calendar occasions.









Demand for all products depends on the characteristics or attributes of the products.

For most food products the prevailing characteristic is to satisfy nutritional needs. Even

though flowers are not essential for survival; they possess other characteristics that are

important to food products and which influence the buying decision. Because flowers are

not essential for survival there is a portion of the population composed of non-buyers or

infrequent buyers. Therefore there is a considerable gap between the decision of buying

or not, and this decision is based upon the demographics of the population and the

occasions or periods. Understanding how consumers make choices whether to buy or not

and the perceptions of the characteristics of the products are essential to understanding

the flower demand (Girapunthong, 2002).

There are three groups of factors that affect the demand for floral products:

external, controlled, and seasonal factors. External factors of demand include inflation,

wages, prices, unemployment rate, demographic factors and other economic variables.

Controlled factors of demand may be used to change perceptions and awareness by

means of promotions, product developments and innovations. Even though demographic

characteristics cannot be changed; the perceptions and behavior of different demographic

groups can be influenced. For example, a promotion program would not change the age

of the consumers, but instead it can target different attributes of the products to influence

purchase decisions by different age groups. Seasonal factors also affect the demand for

flowers. There are certain calendar occasions where the demand for flowers is higher

compared to other non-calendar occasions. The most common calendar occasion dates

are Mother's Day and Valentine's (Ward, 1997).









In order to analyze the demand for flowers, two types of analysis will be made.

First, market penetration will be considered and second we will analyze buyer frequency.

Because flowers are non-essential for survival, in a typical month the percentage of the

population that buys flowers is less than five percent. From this fact arises the need to

understand how consumers make the choice to purchase or not and what the factors are

that influence their purchasing decisions. After determining the factors that affect their

purchasing behavior we can simulate and design specific programs to increase the entry

of new consumers. Once a person becomes a consumer of flowers, and then the

remaining question is the frequency of buying. In an attempt to increase the total

purchases it is also of great importance to understand the factors that influence the

purchasing decisions among consumers of flowers. These two factors will be addressed

in detail in a following chapter including the factors that may affect consumer responses.

Overview of the Industry

Floriculture has been one of the fastest growing sectors of U.S. agriculture. This

sector has had a traditional average annual growth rate of about 5 percent from 1993 to

2003. However, for the first time in two decades, grower sales have remained relatively

flat from 2001 to 2002, with an increase of only 1.6 percent. Total floriculture sales at

wholesale for large growers, that is, growers with sales of one hundred thousand dollars

or more per year, increased to almost 4.9 billion dollars in 2002, up from 3.2 billion

dollars in 1996, which represents an increase of about 54 percent. From this total, there

was an increase of 2.98 percent for fresh cut flowers, 25.25 percent for potted flowering

plants, 22.36 percent for foliage plants, and 73.85 percent for bedding and garden plants

as we can see in Figure 1.1 (United States Department of Agriculture [USDA], 2003).









As shown in Figure 1.2, the number of large growers increased from 4,566 in 1992

to a peak of 5,200 in 1998 and then decreased to 4,741 in 2003 for a total increase of 3.83

percent. Average sales per large grower from 1992 to 2003 increased 57.66 percent for

cut flowers, 47.44 percent for potted flowering plants, 91.59 percent for foliage plants,

and 70.38 percent for bedding and garden plants. Figure 1.3 shows average sales per

large grower by crop group. The total number of growers, including large growers and

small growers, in 1998 was 14,308 with total greenhouse production area of 654 million

square feet. For the 1993 to 1998 period, there was a decreasing trend for the total

number of growers; however, the number of large growers was increasing combined with

an increasing trend of production. From this fact we can see that during this period, there

was a transition, which the production of floral products shifted from small producers to

large growers with higher production. After 1998, the number of large growers also has

had a decreasing trend in combination with an increasing production trend. This decline

in the number of growers has been attributed to increased import competition and

consolidations to achieve economies of scale, such as contract production with large

retail chains (Schumacher et al., 2000).

The value of total U.S. imports of floriculture and nursery products increased from

712.4 million dollars to 1.2 billion from 1994 to 2003 as shown in Figure 1.4. The

countries from which the U.S. imports the most are Colombia, Canada and the

Netherlands for a combined value of 908.8 million or 72.72 percent of total imports in

2003 (Figure 1.5). Cut flowers represented 59 and 49 percent of the imports for 1994 and

2003, respectively. The total value of cut flower imports increased 45 percent from 420

million in 1994 to 611 million in 2003.













Large Grower Sales Millions
1 00


-Cut Flowers -Green House
-Dry/Artificial


1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Figure 1.1 Large Grower Sales by Crop Group From 1985 to 2003. Source. Economic

Research Service, USDA, 2003.


5400 00




5200 00




5000 00




4800 00




4600 00




4400 00


Number of Large Growers


420000
1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003

Figure 1.2 Total Number of Large Growers in the US From 1992 to 2003. Source.

Economic Research Service, USDA, 2003.













Average Sales per Large Grower Millions
080


-Cut Flowers -Green House I
-Dry/Artificial -Outdoor



060 -








00 40







0 20









1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003




Figure 1.3 Average Sales Per Large Grower by Crop Group from 1992 to 2003. Source.

Economic Research Service, USDA, 2003.


Million $


1 60 MImports IExports


140

1 20

1 00

0 80

060

0 40

0 20


0 00
1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002
1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003




Figure 1.4 US Total Imports and Exports from 1976 to 2003. Source. Economic Research

Service, USDA, 2003.
















Others
27.3%


Canada
26.4%


Netherlands
18.5%


Colombia
27.8%


Figure 1.5 US Imports of Cut Flowers and Nursery Crops by Country, 2003. Source.
Economic Research Service, USDA, 2003.




Other
27.3%










Canada Netherlands
52.9% 19.8%









Figure 1.6 US Exports of Cut Flowers and Nursery Crops by Country, 2003. Source.
Economic Research Service, USDA, 2003.









The major exporter of cut flowers to the U.S. is Colombia with 343.6 million in

2003, for 56.21 percent of the total U.S. cut flowers imports. In terms of nursery stock,

the major exporters to the U.S. are Canada and the Netherlands with a combined value of

473.6 million or three quarters of the total U.S. nursery stock imports. Total U.S. exports

of cut flowers and nursery stock increased from 252 million to 272 million from 1994 to

2003. About 53 and 20 percent of U.S. floricultural exports are to Canada and the

Netherlands, respectively, as shown on Figure 1.6 (USDA, 2003).

Even though fresh cut flowers, potted flowering plants and dry-artificial flowers are

fundamentally different and substitutable to some degree, there are certain similarities of

attributes among these products if we analyze them in terms of the purpose of use. They

can be used to express love, thanks, reflect emotions, project beauty, and show

environmental concerns. Consumer expenditure patterns may change among these

products even though they are physically different. These consumer patterns are affected

by many factors, including income, purpose of use, occasions, information, perceptions,

and sources for purchases. The level of consumer expenditures depends on three basic

components: market penetration, frequency of transactions among buyers and prices

(Girapunthong, 2002). Demand analyses for floral products differ among other

agricultural commodities in the sense that for other agricultural commodities the quantity

consumed is used directly in the analysis. In the specific case of flowers, a consumer may

purchase a single stem rose or an arrangement. Therefore demand studies on flowers

generally replace quantity observed by the number of transactions given in a defined

period of time.









Given the reasons why consumers buy flowers and the fact that flowers are non-

essential for survival, one can observe that the consumption of floral products has

significant fluctuations over time. These fluctuations result in a higher number of

transactions during special calendar occasions during the year. The most important

calendar occasions are Valentine's Day, Mother's Day, Easter/Passover, Thanksgiving,

and Christmas/Hanukah. The American Flower Endowment consumer tracking study of

1999-2000 reported that Christmas/Hanukah had the highest number of transactions, and

represented one-third of all calendar's occasions; Mother's Day, Easter/Passover,

Valentine's Day and Thanksgiving accounted for 20, 18, 16 and 5 percent respectively. It

is of vital importance that any demand analysis for flowers takes into consideration these

special calendar occasions.

In the last decade, the industry has experienced many changes, including industry

programs adopted to increase the total demand for flowers. Brand and generic programs

have been adopted to entice the demand for floral products. The fundamental difference

between brand and generic advertisement programs is that brand advertisement is

adopted by a firm to benefit specifically that firm while generic advertising seeks to

increase the demand of the whole flower industry. Examples of generic advertisement in

the flower industry include PromoFlor and the Flower Promotion Organization (FPO)

promotion programs. PromoFlor was a generic promotion program implemented in 1993,

which had the main objective to expand the total demand for fresh cut flowers and

greens. Ward (1997) showed that the PromoFlor program was successful in increasing

the expenditures per buyer, generating a net gain of 5.6 dollars of additional revenue for

every one dollar spent on the promotion program at the wholesale level. However,









PromoFlor was terminated in June 1997. Rimal (1998) explained that one possible reason

for the termination of the program was equity concerns regarding the distribution of

benefits among the fresh cut flower system. The FPO implemented a promotion program

that targeted five U.S. cities with the goal of increasing the frequency of buying fresh cut-

flowers by existing female flower buyers for non-calendar occasions. Potential changes in

market penetration and buying frequency were adopted as measurement criteria for

judging consumer responses to the promotions. The program had a positive impact on

attracting new buyers and increasing buyer expenditures per transaction. For every dollar

spent on the promotion program there was an additional $9.5 dollars generated on

expenditures of fresh cut flowers in the target area (Ward, 2004).

In addition to promotion programs, the development of new technologies, such as

the Internet, has made possible the creation of new sources for buying floral products.

Examples of Internet-based firms include FTD, 1-800-Flowers, and Teleflorist. These

businesses have created product diversification such as floral baskets and bouquets,

which can influence the purchasing decisions of consumers and their tastes and

preferences. The main objective of these firms is to increase the demand for floral

products through the use of technology and make the services more convenient

(Girapunthong, 2002).

Problem Statement

The floriculture industry is one of the fastest growing sectors of agriculture in the

U.S. In 1996, floriculture ranked seventh among commodity groups, behind only cattle

and calves, dairy products, corn, hogs and soybeans. Floriculture crops, defined as cut

flowers, cut cultivated greens, potted flowering plants, potted foliage, bedding and garden

plants, accounted for about one third of grower cash receipts for floriculture and









environmental horticulture (Schumacher et al., 2000). In order to continue this growing

trend it is of vital importance that one obtains insight into consumer preferences on floral

products. Specifically, there are two key factors that need to be analyzed in order to

understand how consumers base their decision to buy or not to buy floral products:

market penetration and buyer frequency. Understanding what the factors are that

influence non-buyers of floral products to become buyers, and the factors that influence

buyers to increase their expenditures on floral products, is vital information that the

industry can use to design specific programs targeting different demographic groups

according to their specific preferences on flowers.

Objectives

The general objective of this study is to analyze the factors that drive the demand

for flowers in the U.S. in terms of market penetration and frequency of buyers for cut

flowers, potted flowering plants, dry/artificial and outdoor plants. Three specific sub-

objectives will be accomplished in order to achieve the overall main objective:

1. Given the number of buyers and households, analyze the factors that attract non-
buyers of floral products to become a buyer, which is market penetration.

2. Examine the factors that contribute to increasing consumer expenditures on flowers
depending on the type of product, source, reason for buying, seasonal
considerations and demographic characteristics.

3. Use the results from the market penetration and buyer frequency models to make
simulations on specific combinations of the product attributes and demographic
characteristics, to rank the importance of those factors impacting both market
penetration and frequency.


Research Methodology

Our data set shows the number of buyers and the number of households, which

would, allows one to calculate the market penetration ratio. Also the data have showed









total expenditures, number of buyer transactions and quantities. As stated before, in the

case of flowers, it is not of much advantage to use the quantity of purchases because that

it is hard to record whether a quantity of one means one single stem rose, or an

arrangement of multiple flowers. That is why the number of transactions is replaced by

the quantities. The frequency of buying can be calculated by dividing the number of

transactions by the number of buyers. Also price can be calculated by simply dividing the

expenditures by the number of transactions. The focus of the study is divided into two

main parts: market penetration models and buyer frequency models.

Based on these main variables mentioned in the first paragraph, market penetration

models are used to analyze what factors influence consumers to become a buyer or not.

This model is one of the main two topics of this study; the second part will address buyer

frequency models. When a consumer has become a buyer, that is, households with at

least one transaction, what are the factors that influence how much consumers buy.

Because both models, market penetration and buyer frequency, have a cluster of

observations on the lower limit, a model is needed that will take into account its

asymptotic distribution. The market penetration model has a lower limit at zero, while the

buyer frequency has a lower limit of one, since in order to be defined as a buyer a

household must have made by definition at least one transaction per month or more. The

model that deals with this type of clustering of the data is the Tobit model. A Tobit model

combined with frequency of buying models will be used in order to analyze the factors

that affect the number of transactions in a given period of time.

After the implementation of market penetration and buyer frequency models, the

results can be used to simulate different products, reasons for buying, outlets sources and









demographic characteristics such as gender, age, income groups, etc. These simulations

will result in specific recommendations for the flower industry in an attempt to increase

the overall demand for flowers in the U.S.

Data and Scope

Data used in this study are aggregate data collected by the National Panel Diary

Group (NPD) and sold through the American Flower Endowment. The data provide

statistics on behavior of consumers including transactions and expenditures on flowers

for both gift and self-use. NPD data were available from consumers purchasing diary

completed by households from a large demographically representative sample of U.S.

households. The data have two separate variables for number of households and the

number of households that buy flowers. This separation between household buyers and

non-buyers would allow the calculation of the market penetration ratio, or the percentage

of the overall sample that are buyers of floral products. Monthly purchasing data from

July 1993 to June 2004 will provide information regarding the number of households

buying flowers, the total number of households, expenditures, number of transactions and

quantity for both gift and self-use of flowers.

Flower data used in this analysis are categorized into four different income groups:

under $25,000; $25,000-$49,999; $50,000-$74,999; and $75,000 or more. These income

groups have data from five main categories: product form, purpose of the purchase, time,

demographic characteristics, and geographic location. Product form refers basically to the

four sub-categories: cut flowers, potted flowering plants, dry/artificial and outdoor; also

as subcategories of cut flowers there are arrangements, single stem/ bunches and others.

The purpose of the purchase is either for self-use or for gifts. The time of the purchase

(calendar vs. non-calendar occasions) is recorded along with demographic characteristics









including income, education, employment, occupation, family size, gender, marital status

and regional location.

Organization of the Study

This study has seven chapters. Chapter 2 is a review of literature on demand

theory, floral demand analysis and literature on market penetration and buyer frequency.

Chapter 3 consists of descriptive statistics of consumer demand for the flower industry in

the U.S. with expenditures, transactions, number of households, reasons for buying,

product form and demographic characteristics being addressed. Chapter 4 presents a

conceptual framework for market penetration and frequency theory and measurement.

Chapter 4 is basically where econometric models and the development of estimation

techniques will be constructed. Chapter 5 includes the model specification and

estimation. In this chapter the results are discussed and interpreted. Chapter 6 is

simulation analysis and has two main sections: (1) ranking of market penetration

variables and ranking of frequency of buying; and (2) projections: using demographic

trends. Finally, Chapter 7 gives a summary of the study.














CHAPTER 2
LITERATURE REVIEW

This chapter consists of four sections. First, the consumer behavior will be

addressed with an emphasis on preferences and choices. The properties of preferences

will be listed and described. Indifference curves, marginal rate of substitution and utility

maximization will be discussed. Second, consumer demand analysis will be addressed

and discussed. The consumer allocation problem and demand properties will be the main

focus of this section. Third, different marketing models will be presented and discussed.

These models are the fundamental theory behind market penetration models and buyer

frequency models. The two main parts of this section are: market penetration models,

where different functions for market penetration models will be described verbally and

mathematically; and frequency of buying models, where we will introduce the notion of

repeat-buying and will describe the model used to analyze buyer frequency models. And

finally, past studies on the flower industry will be listed and summarized.

Consumer Behavior

Consumer behavior is often presented in two ways: in terms of preferences or in

terms of possibilities. In discussing consumer behavior, generally the main focuses are

preferences, axioms of choice and utility functions and their properties. Unlike

preferences, choice as an opportunity is often directly observable so that, to the extent

that variations in behavior can be traced with variations in opportunities, one has a

straightforward and objective explanation of observed phenomena.









Preference is a key factor in consumer or buyer behavior. Preference determines

choices made by buyers of most products and services within the limits of a set of defined

constraints. Buyer preference will determine what goods or attributes of a good are

selected and will also determine the selection of one good over another. An

understanding of this concept is vital to the market development of a product such as

flowers, a product consumed for its aesthetic value.

Considering all factors, when a consumer reports that "A is preferred to B", then

that particular consumer feels better under situation A than under situation B. This

preference relation is assumed to have three basic properties: completeness, transitivity

and continuity. The completeness property states that if A and B are any two situations,

the individual can always specify one and only one of the following possibilities: (1) A is

preferred to B, (2) B is preferred to A, or (3) A and B are equally attractive. Individuals

completely understand and can always choose the desirability of any two alternatives A

and B. The transitivity property infers that if a consumer reports, that "A is preferred to

B" and that "B is preferred to C," then the consumer must also report, that "A is preferred

to C." Therefore, the transitivity property assumption states that individual choices are

internally consistent. The continuity property indicates that if an individual reports, that

"A is preferred to B," then under similar accommodated circumstances, the consumer

must also report that "A is preferred to B" (Nicholson, 1998).

Given the assumptions of completeness, transitivity and continuity, it is possible

to establish that the consumers are able to rank order all possible choices from the least

desirable to the most desirable. The ordering implies an underlying level of utility or

satisfaction that is derived from each choice. If an individual has a preference of choice A









over choice B, then the utility obtained from A, denoted by U(A), is greater than the

utility derived from B, U(B). Utility can be expressed in ordinal number values, where

the higher the value the higher the utility. These values will reflect the preference

ordering for a set of choices.

Because utility refers to overall satisfaction from the consumption of a product or

a service, it is influenced by a variety of factors. A consumer's utility is influenced by the

following factors: consumption of physical commodities, income, peer group pressures,

personal experiences, and the general cultural environment (Nicholson, 1998). Individual

preferences or utility are assumed to be represented by a function of the form

(2.1) utility = U(X,,X2,X3,...,X,; other),

where X, ,N refers to factors or product attributes and the other are variables that affect


the utility for the product such as demographic characteristics.

A consumer preference mapping can be shown graphically through the use of

indifference curves. An indifference curve represents all combination of product

attributes that provide the same level of satisfaction (utility). For simplicity, one can

assume that there are only two product attributes for a product, attribute A and attribute

B. Then the utility for the product would be represented as U(A,B). Thus an individual is

indifferent among the combinations represented by the points graphed on the indifference

curve (Pindyck and Rubinfeld, 2001). Figure 2.1 is a graphical representation of

consumer preference mapped by an indifference curve.

In the figure, there are only two products A and B. For a particular consumer, the

same level of utility is attained at both points A and B (note those are points along the

same indifference curve and therefore provide the same level of utility. Hypothetically a









consumer would be willing to give up some utility obtained from product A in order to

increase the utility received from product B, and vice versa. In other words, it is

suggested that a consumer would be willing to receive a product with a less favorable

amount of B in order to obtain more quantity of product A, or a consumer would be

willing to accept less of product A in order to obtain more of product B.




Comb B





/ Increasing
VUtility


B2


A
B*


Budget
line

A2 A*
Comb A

Figure 2.1 Graphic Presentation of Indifference Curves

This situation is true if all other things remain constant (ceteris paribus

assumption). An individual receives the same total utility by consuming the combination

A*,B* (point A) or the combination A2,B,, (point B). Again, the curve represents all the

consumption bundles that provide the same level of utility or stated in a different manner,

all the bundles that the individual ranks equally in terms of derived utility. However the

consumption would take place at point A because given the budget line that would be the









point where utility would be maximized and the budget constraint would hold. The slope

of the indifference curve is negative, meaning that if an individual is willing to give up an

amount of product A, then he must be compensated with an increase in the amount of B

in order to remain indifferent between the two bundles. This compensation represents a

trade-off between products A and B. This negative slope of the indifference curve is

called the marginal rate of substitution (MRS) at that point, and it is expressed as follows:


MRS= dU I
(2.2) dB



where the notation means that the slope has to be calculated along I, indifference curve.

The slope or the MRS represents the trades a consumer would voluntary make. Utility

increases as the indifference curve shifts out due to other constraints being relaxed. This

fact can be seen in Figure 2.1, as U(I,) < U(I2) < U(I3). In order to maximize the

overall utility, a consumer would be located on the indifference curve situated as far from

the origin as possible within the context of the constraints the consumer faces. In Figure

2.1 the budget line establishes the combination of products along the indifference curve.

The budget line is the constraint that a consumer faces as a result of relative prices or

costs and a fixed level of income. In other words, the budget line, which is simply the

ratio of the prices of product A and B, represents all combinations of A and B for which

the total amount of money spent is equal to consumer income. Consumers maximize

utility by choosing a bundle that is on the indifference curve and that is tangent to the

budget line.









Consumer Demand Analysis

There exists extensive literature in demand analysis. Some of the early works

documented on demand analysis include Hicks (1946) and Samuelson (1947). Other

demand analysis studies that are often used for pedagogic purposes include Goldberg

(1967), Phlips (1974), Powell (1974), Theil and Gabrielson (1975, 1976), Barten (1977),

and Deaton and Muellbauer (1980). The formulation of the utility maximization problem

implies the existence of a utility function with specific properties. The purpose of demand

analysis is to study from the utility function to the demand function and analyze the

properties these functions have (Johnson et al., 1984).

Utility Maximization and Demand Functions

The previous section explained the assumptions about consumer behavior. These

assumptions are introduced into consumer demand analysis through the specification of a

utility function. The utility function measures the level of satisfaction a consumer would

obtain from a product or service on a given period of time. The utility function is denoted

by the function:

(2.3) u = u(q,)

where the expression means that the utility is a function of the levels of consumption at

period i. The utility is represented by an ordinal number with no definite scale; the higher

the number the higher the level of utility obtained from the consumption of a bundle of

goods or services.

The utility function defined in 2.3 has three basic assumptions. First, it is assumed

to be strictly increasing. A function f(x) is said to be strictly increasing on the interval

(a,b) if









(2.4) f(x,) > f(x,) whenever x, > x,.

Second, it is assumed that the utility function is quasi-concave. A function f(x) is quasi-

concave over the interval (a,b) if

(2.5) [Ax, + (1- A)x,] > min[f(x,), f(x,)] for all x,, x, in (a,b) and all 0 _< A <1.

The function f(x) is strictly quasi-concave if the strict inequality (>) holds. Third, and last

a utility function is assumed to be twice continuously differentiable. A function f(x) is

continuous at the point x=a if the following conditions hold:

(2.6) C f(x) and f(a) exist and C f(x) = f(a)
x->a x->a

The interpretations of these conditions are easier if the indifference curve is used (Figure

2.1). Strict quasi-concavity of the utility function (2.3) ensures that the indifference curve

does not contain linear segments or bends back on themselves. The assumption that the

utility function is strictly increasing implies that a consumer prefers more to less. The

twice continuously differentiable assumption assures that the indifference curves are well

defined and not kinked.

Marginal utilities are the first partial derivatives of the utility function. These

marginal utilities should be interpreted as the increase in total utility that takes place

because of the consumption of an additional unit of the commodity per unit of time.

These derivatives of the utility function are represented by:


(2.7) =- > 0, (i=1,2,3,...,n),


and they are positive because of the continuity and the differentiability assumptions; the

second partial derivatives of the utility function are, by Young's theorem, symmetric, and

they are represented as










(2.8) u = = (ij = 1,2,3,...,n).


The second partial derivatives of the utility function can be interpreted as the rate of

change of the first partial derivatives and therefore the behavior of the marginal utility for

each of the commodities. In order for the analysis to be complete it must include how the

marginal utilities change with changes in the consumption levels of other commodities,

that is, u0, when i # j.

The utility function (equation 2.3) is maximized subject to a budget constraint,

that specifies that consumers have no savings, or in other words, that consumers spend all

of their income:

(2.9) p'q = m,

where p = (p,) is the n-element column vector of the prices and m is consumer income.

Hence, our maximization problem is:

(2.10) Max u = u(q,)

Subject to: p'q = m .

The maximization of the utility function subject to a budget constraint in equation (2.10)

is carried out by the Lagrangian method. According to this method the Lagrangian

expression is formed as follows:

(2.11) L(q, A) = u(q)- A(p'q- m)

where A is the Lagrangian multiplier and it is interpreted as the marginal utility of

income. Differentiating the Lagrangian equation (2.11) with respect to each of the

arguments, q, and A yields the first order conditions (FOC):

(2.12) U Ap = 0, and p'q m = 0,









where uq is the vector of derivatives of the utility function with respect to the quantities,

q,, (i = 1,2,3,...,n).

The second order conditions of the Lagrangian for a maximum can be expressed in

the following manner:

(2.13) x'Ux < 0 for all x such that p'x = 0.

The matrix U is the Hessian matrix, and condition (2.13) is assured because of the

assumption that the utility function U is strictly quasi-concave. The system of equations

can be solved for ql,..., q and A in terms of prices and income. The resulting unique

expressions are:

(2.14a) q, = q, (pl,..., p,,m)

(2.14b) A A(p,...,p",m).

The demand function q, is of extreme importance in both theory and practice, because it

describes how a consumer will behave when faced with an alternative set of prices and a

particular income. The term A shows that the marginal utility of income depends on both

prices and income.

Economists often find it more convenient and useful to express the restrictions on

the demand system in terms of elasticities rather than derivatives (Johnson, et al, 1984).

Elasticity of demand can be calculated from the maximization procedure described

above. There are three basic elasticities of demand: (1) own-price elasticity, (2) cross-

price elasticity, and (3) income elasticity. The own-price elasticity of demand can be

defined as the proportionate change in the quantity purchased to the proportionate change

in its own price as follows:









tq, P
(2.15) O,, <0.
Op, q,

If s,, =1, then there is a unitary elasticity of demand, which means that a 10 percent

increase in price will decrease the quantity consumed on the same proportion, in this case

10 percent. If the elasticity of demand is greater than one in magnitude (absolute value),

then the demand is price elastic, because the percentage decline in quantity demanded is

greater than the percentage increase in price. If 0 < ,, < 1, then, the demand is inelastic.

The cross-price elasticity of demand of a good is the responsiveness of the

quantity of that good to changes in prices of other goods:

5q, p,
(2.16) E q
Op, q,

If the cross-price elasticity of demand is negative (e, < 0), then good i and good j are

complements, while if the cross-price elasticity of demand is positive (e, > 0) then the

two goods are substitutes.

The income elasticity of demand can be defined as the proportionate change in

quantity purchased relative to changes in income, given that prices are held constant:

5q, m
(2.17) rm -
cm q,

If the consumption of a good increase with income increases, that is the income elasticity

of demand is positive, then these types of goods are defined as normal goods. If the

marginal propensity to spend on q, is less than the average propensity to spend on q,

(0 < rm < 1), then these types of goods are referred as necessity goods. If the marginal









propensity to spend on q, is greater that the average propensity to spend on q, (7r, > 1),

then the good is known as a luxury good.

Up to this point the maintained hypothesis of demand theory is that a consumer

will select from the set of affordable commodity bundles the one that would yield that

consumer the maximum possible utility attained. The assumptions and results of this

hypothesis have been explained above in the utility maximization framework. A different

approach to analyzing the consumer optimization problem can be used by applying the

duality concept. Duality was introduced by Hotteling (1932) and developed and

popularized by Shepard (1953) with his work on production and cost functions and

employed to advantage in the analysis of consumer demand.

The duality formulation of the consumer allocation problem is developed from the

expression:

v(p,m) = maxu(q)
(2.18) q
Sto : p'q = m

where v(p,m) is the indirect utility function. The indirect utility function is defined as the

maximum attainable utility level for a given set of prices and a particular income. The

indirect utility function can be used to obtain the direct utility function. If the indirect

utility function v(p,m) is minimized with respect to prices and income and subject to the

budget constraint, the direct utility function u(q) is obtained. This property of the direct

and indirect utility function is of great importance for analytical purposes. In order for

this relationship to exists, the indirect utility function v(p,m) has to have the following

properties: (1) continuous, (2) decreasing in prices, (3) increasing in income, (4) strictly

quasi-convex in prices, and (5) homogeneous of degree zero in prices and income.









The advantage of using this approach in analyzing the consumer demand problem

is that it is easier to derive the demand functions and associated conditions. For instance,

the uncompensated (Marshallian) demand function for the ith commodity can be obtained

by differentiating the indirect utility function in (2.18) with respect to prices and income

and applying Roy's identity:


[av(p, m)/ m]

In a similar way the uncompensated demand function can be obtained by applying the

Hotteling-Wold identity to the direct utility function (Johnson, et al, 1984):

(au / aq, )m
(2.20) = (q, m) = p, (i= 1,2,3,...,n).



Properties of the Demand Functions

Restrictions on consumer demand functions or properties are obtained from the

manipulation of the first order conditions (FOC) specified in (2.12). Properties of demand

are developed by considering the consequences of shifts in the first order conditions with

respect to prices and income. The results of these shifts are described by the partial

derivatives of the first order conditions for both prices and income. There are four basic

demand properties or restrictions: (1) adding-up restriction, (2) symmetry restriction, (3)

homogeneity restriction and (4) negativity restriction.

Adding-up restriction

The adding-up restriction of the demand function is derived from the

monotonicity assumption of preferences and the budget constraint. The adding-up

restrictions has two main components known as the Engel and Cournot aggregation

restrictions. The Engel aggregation restriction relates to income elasticities and it is









represented as the sum of the income elasticities weighted by its expenditure share,

w, = P, /m.

(2.21) -wi7,m = 1


The Cournot aggregation restriction states that the sum of the cross-price elasticities

weighted by its expenditure shares has to equal the negative of the expenditure share of

good j:

(2.22) w j = -W where j=1,2,3,...,n.


Symmetry restriction

The symmetry restriction states that the matrix of compensated cross-price

substitution effects is symmetric. This is true because of the continuity and

differentiability assumptions of the utility function that follow Young's theorem.

(2.23) W, + WWj,, = wjCj + WJW7rjm, or zj = zjj for i # j,

where E, is the cross-price elasticity, w, is the weighted expenditure share of good i, 7,m

is the income elasticity for good i, and z-, = w,e + w,wJr,m is the Slutsky coefficient

for good i and j.

Negativity restriction

The matrix conformed by the second derivatives of the expenditure function, by

Shephard's lemma, constitutes a substitution matrix and hence should be a negative semi-

definite. For the necessary condition, the diagonal elements of this matrix would have to

be negative, which can be reformulated in elasticity form by using the Slutsky equation

and setting i=j as well as multiplying by p, /q,:

(2.24) E,, + w,7r < 0









Homogeneity restriction

The homogeneity restriction of the demand function states that the sum of all the

direct and cross-price elasticities for a particular commodity i, is equal to its ith income

elasticity:

(2.25) C, = -7,,m where i=1,2,3,...,n.


Marketing Research Models

Marketing is the process by which systems of consumption are developed and

changed by individuals and organizations. A marketing system consists of organizations

that provide goods and services to consumers, the flows of information between the

organization and the consumer, and the physical flow of goods and services (Parsons and

Schultz, 1976). The main focus on marketing research aims to answer questions in regard

to the effects of marketing instruments and specific household demographic

characteristics on various marketing performance measures. Some examples of these

marketing measures include sales, market shares, brand choice and inter-purchase times.

With the results on these measures it is possible to select which marketing instruments

will be used targeting specific household groups. In recent years there have been major

advances in the data collection methods, allowing the researchers to access large

databases on individual consumer habits. In addition to this, through the use of

supermarket loyalty cards and scanners researchers can track what individuals purchase;

hence the researcher gains information not only on stated preference, that is what the

consumer reports, but also revealed preference, that is what the consumer actually

purchased. The large amount of marketing data implies that simple graphical tools and

elementary modeling techniques in most cases are not sufficient to analyze and explain









present day marketing problems (Franses and Paap, 2001). In order to obtain results that

are accurate with reality researchers need to use the available information to build

marketing models that explain and adapt to real life. This research project will use two

marketing models: market penetration models and buyer frequency models.

Market Penetration Models

The market penetration process refers to individuals that are non-buyers of a

product or service and then when given some type of information-based decision they

become buyers. The main question on market penetration is what are the factors that

attract new buyers for the product. In the case of flowers, there has been a change in the

trend of flowers. Advertising campaigns have shown that flowers are not only a product

for gifts, but also for self use for decoration purposes and personal enjoyment. One can

use the available data that includes reasons for buying, transactions, prices, time of the

year (calendar vs. non-calendar occasions), and demographics, to segment the population

and identify the factors that most likely attract non-buyers of floral products to become

buyers. There are many mathematical functions used on market penetration models; this

section will present and discuss the most important mathematical functions used in

market penetration. The main source for this section was the market penetration functions

developed by Fleck (1981).

Logistic function

The logistic function can be expressed in the following differential equation:

(2.26) y' (t) = y(t)[ y(t)]

Sy(t)
where y' (t) = is the increase in demand at period t,
9t


is the demand accumulated at period t,


y(t)









3 is the positive factor of proportionality,

y limit of saturation (y > 0).

The interpretation of this differential equation is quite simple. The increase in demand

y'(t) is determined by the market potential [y y(t)], which is not exploited yet at time

period t and by the stock y(t) reached in t. At the beginning of the penetration process the

demand increase is small, because even though there is a high potential market, few

consumers will accept the product. With the increasing number of buyers, the potential

market decreases and y'(t) also increases. Then when there is a large enough portion of

new buyers of the product, y'(t) decreases because the potential market becomes

smaller.

With a positive demand at period t, the solution to the differential equation (2.26)

can be obtained by the separation of the variables and decomposition of a fraction into

partial fractions:


(2.27) y(t) = 1 -bt
1+e

One of the criticisms of the logistic function is that it describes the process of market

penetration as an endogenous process. Because of this fact, the parameters are constant

meaning that they are not affected by external market influences.

The Pyatt function

This model combines both exponential and logistic growth. Exponential growth

with limit saturation and with no initial resistance to buy describes products that are

frequently bought (routine purchases). Similarly, this situation also applies to purchases

caused by new trends, where at the beginning of each period the targeted consumer









groups are already well informed about the product and the desire to buy increases. The

foundations of the Pyatt model are based on the following differential equation:

(2.28) y' (t) = A(y y(t)) + gy(t)(y y(t)).

The first term is the exponential growth and the second is the logistic growth. Depending

on the value of A the shape of the curve will be different. If A = 0, then the logistic

function is obtained. If A <1, the logistic behavior is dominant. If A =1, then a

hyperbolic function is obtained. The Pyatt curve is obtained as the solution of the

differential equation (2.28) with the initial condition y(0)=0 :

1 e (l+Z)t
(2.29) y(t) = y.
1 e (1+Z)t


The Gompertz function

This function was first applied by the economist Gompertz in 1925 for the

analysis of mortality. The Gompertz function can be expressed as follows:

(2.30) y(t) = yebc

where both b and c are constant and b>0 and 0
saturation y = e", then

(2.31) y(t) = eabc

The Weblus model

Weblus first developed this model in 1965 while studying the spreading of luxury

goods. Weblus used as examples of luxury goods washing machines, refrigerators,

television sets, etc. He did research on market penetration of televisions and automobiles.

He assumes that the purchasing power and interests are distributed differently in society.

The result of this is that the number of purchases decreases over time. This is due to the









fact that the penetration process slows down as the degree of saturation increases. The

above Weblus model can be expressed in an equation of the form:


(2.32) y'(t) = y(t)(1- y(t)).
t

Here the constant term 3 described on the logistic model (2.26) is replaced by the time

dependent parameter a/t. In this equation, Weblus uses as the limit of saturation (y) the

relative limit of 1, which corresponds to the total market or a market share of 100 percent.

1
The solution to the differential equation (2.32) where y(to) = reads:
2


(2.33) y(t)

1+


where to is the period of time between the beginning of the penetration process and the

point when the saturation reaches one half or 50 percent of market share is attained.

The log-inverse function

Prais and Houthakker applied this function for the first time when analyzing

family budgets in 1955. This function is defined by a growth rate, which is determined in

the following manner by the stock level y(t) and time squared:

1
(2.34) y'(t)= by(t)


where b is a positive constant, and then a solution is obtained where a = In for t oo

b
a(2.35) y(t)
(2.35) y(t)= e









The disadvantage of the log-inverse function is that it only possesses one degree of

freedom or in other words, one parameter, and then it can be difficult to adjust this model

to real data.

The lognormal function

The logistic and the cumulative normal distributions are very similar. They both

possess a disadvantage in that both have a property that one half of the limit of saturation

is obtained at the point of inflection. Then, in order to obtain a function that is similar but

without that property the lognormal function was defined as follows:

t 12 1(og -
(2.36) y(t) = yJ 1 e dr,


where a is the standard deviation and p is the mean of the distribution.

Repeat Buying Models

Repeat buying is when a consumer buys a product more than once in a given

period of time. Consumers have pre-purchase needs, perspectives, attitudes, the

experience of previous usage, and external influences, such as advertising and promotion

programs, retail availability, personal selling and word of mouth effects, and differences

in products, services and prices. The consumer has to make decisions regarding what

products to buy and at what prices and where to buy the products. All of these

characteristics form a post-buying experience in the customer's mind after the purchase

takes place; based on all these factors a consumer would choose depending on the level

of satisfaction or utility obtained from the product or service whether to re-purchase the

product or not. There are basically three cases of repeat buying situations that can be

defined. First, if a consumer buys more than one product in one or more purchase

occasions in the given time period. In this case, consumers differ in how often they repeat









buying the products. The frequency of buying would be 0 for a consumer that did not

purchase the product and 1 for consumers that purchased the product once. For the repeat

buyers the frequency will be 2, 3, 4, etc., depending on the number of repeat buying

occasions they purchased the product. The second way of repeat buying refers to

consumer that may buy the product in more than one time period. Then a model can be

formulated for repeat buying behavior under stationary and no trend conditions. The third

and last form of repeat buying behavior is that more than one unit may be purchased on

the same purchase occasion (Ehrenberg, 1988).

In the case of the data set, the data are organized in such a way that we do not

have information about the non-buyers. The frequency of buying is then the number of

purchases a household made in a given period of time (months). The frequency of buying

of flowers is affected by external seasonal factor. As an example, the frequency of buying

as well as the total number of buyers increase during special calendar occasions such as

Mother's Day, Valentine's, Christmas, etc. In order to analyze the frequency of buying,

and because of the fact that the data set does not contain any information about non-

buyers of floral products, we then will employ the Tobit model.

The Tobit model is an extension of the Probit model. The Nobel Prize winner

James Tobin developed the Tobit model in 1958. Because the only available data is on

consumers that did purchase floral products, then the Tobit model will be the most

appropriate model to make the analysis. Consumers are divided in two groups, one

consisting of n1 consumers about whom we have information on the regressors (income,

education level, gender, etc.), as well as the regressand (expenditures on flowers); and the

second group, n2 consists of consumers about whom we only have information about the









regressors. A sample consisting of people on the second group is known as a censored

sample. That is the reason why the Tobit model is also known as censored regression

models or a limited dependent variable model (Gujarati, 1995).

The general mathematical formulation of the Tobit model can be expressed in the

following equation assuming the constraint from below:

(2.37) y, = p' x, + ,,

y, =0 if yI <0

y, = y, if y, >0.

There are three different conditional functions that can be analyzed depending on the

purpose of the study. For the index variable, sometimes called latent variable, E[y,'] is

/8'x,. However, if the data are always censored this result would not be useful. Then for

any observation randomly drawn from the sample, which may be censored or not, the

expected value is:


(2.38) E[y, x,] = o 1('x, + A,),



where, A, =( / )
( 'x, / )

The estimation of the model is done through the use of the maximum likelihood

technique. The log-likelihood function for the censored regression model is:

1 y x'x,
(2.39) logL= 1 2 log(2 r)+log + -1' 2 + log 1-
y,>0 2 C o 0 1

The two parts correspond to the classical regression for the non-limit observations and

the relevant probabilities for the limit observations, respectively. This log-likelihood









function can be simplified following Olsen's reparameterization, where y= p//o and

0 = 1/to, to obtain:


(2.40) logL = 1[log(2r)- log +(0y, 'x,)2]+ log[1- (y'x,)].
y,>0 2 y,=0

Review of Past Studies on Flower Products

The floriculture industry has evolved rapidly in recent years. The introduction of

mass-market retailers such as supermarkets, department stores and Internet-based

business has changed the marketing paradigm of floriculture. Compared to the other food

products such as milk, meat, citrus, etc., floriculture lacks an extensive marketing

literature. Early studies focused on consumer preferences toward fresh cut flowers. The

following section presents some studies reported on floriculture that will be relevant to

the analysis of this research.

Miller (1983) performed an extensive sub-sector analysis for the fresh cut-flower

industry in the United States. The methodology employed by Miller was to use an

approach that followed the structure, conduct and performance framework in order to

analyze the existing conditions of the industry and to predict future trends. In his analysis,

Miller observed that there were special calendar occasions when the demand for flowers

was substantially higher, and other non-calendar occasion where the demand was

decreased substantially. He also determined that the demand for flower arrangements was

inelastic, meaning that consumers are not highly responsive to changes in price of flower

arrangement products. Also Miller observed and reported a rapid growth of mass-markets

in the fresh cut-flower industry and predicted that they would become more important in

the future.









Tilburg (1984) in his study titled "Consumer Choice of Cut Flowers and Pot

Plants" analyzed consumer panel data of households in the Netherlands. The main

objective of the study was to relate aspects of consumer behavior on cut flowers and pot

plants to marketing variables and demographic characteristics of households, to

determine whether market segments exists or not, and to determine the feasible

application of marketing models in the flower industry. He identified three market

segments: the first segment consisted of 44 percent of the households and was sensitive to

prices but insensitive to national advertisement; the second segment consisted of 40

percent of the households, and was insensitive to both prices and advertisement; and the

third segment, with 13 percent, was sensitive to prices and advertising. All of the

segments were separated by these factors as well as demographic characteristics of the

households. He estimated the mean price elasticity of demand for cut flowers and pot

plants to be -0.28 for non-habitual buyers, and -0.81 for habitual buyers.

Behe (1989) in her study "Floral Purchase Behavior of Pennsylvanians" analyzed

the consumer purchasing behavior in the flower industry at the retail level. She

recommended three ways to segment retail flower markets based on buyers attitudes

toward flowers, income and demographic characteristics of the Pennsylvania households.

The first type of segmentation was segmentation by product, which basically

distinguished between fresh cut flowers and potted plant customers. The second type of

segmentation was segmentation by volume of purchase, where she had three categories:

light, medium and heavy consumers. The third and final segmentation was by location of

the purchase, which included supermarkets or florists.









Behe et al. (1992a), carried out an analysis of consumer purchases of floral

products in Ohio supermarkets. They used a mail-survey instrument to identify the

factors that influence purchases of floral products in supermarkets. They used a principal

components analysis and identified 34 independent factors that accounted for 64 percent

of the variation. These factors were grouped into five different categories as follows:

product, consumer, store, use (gifts vs. self) and use (location). These factors represented

the most important influences on floral buying decisions and were used to define five

different market segments of supermarket floral customers.

In the same year, Behe et al. (1992b) carried out a follow-up study titled "Market

Segmentation of Supermarket Floral Customers." They used the same data set obtained

from the mail survey instrument and identified the 34 most important factors affecting

floral buying decisions. They applied cluster analysis on survey responses to create five

homogeneous consumer segments. There were 14 factors that contributed to most of the

differences among segments, including factors of product assortment, number of

purchases, degree of personal use, and packaging importance. The five clusters or

segments were: (1) friendly buyers, which composed 20 percent of total customers; (2)

married man, which also had 20 percent of total customers; (3) Selfers, with 30 percent of

the sample; (4) annual buyers, with 25 percent of the total consumer sample; and (5)

educated mothers with only 5 percent of the total sample. The importance of this study is

the finding that clusters can be used by supermarkets and florists managers to target

different potential market segments.

Becker (1993) in his study "Products, Services, and Consumer Perceptions of

Service Quality in the Retail Floral Industry of Texas", presented differences in service









quality between supermarkets and florists in the Texas region. The differences in the

types of retail outlets were based in terms of types of products sold, custom design and

other in-store services, delivery options and convenience. Becker found that there were

four main differences between florists and supermarkets. First, florists carried out more

non-perishable items than supermarkets. Second, florists provided more custom design

compared to supermarkets. Third, almost all florists (99 percent) had delivery services

compared to only 41 percent of supermarkets. And fourth, the feature convenience was

better perceived at supermarkets, because often they open more days of the week and for

longer hours.

Ward (1997) evaluated PromoFlor's impact on the demand for fresh cut-flowers

and greens in the U.S. using household data on flower consumption. Ward suggested that

changes in the flower demand occurred because either change of the number of buyers

from period to period with the entry of new buyers that increase the number of buyers of

floral products, or changes in purchases per buyer. Ward found that the impact of

PromoFlor was different between the four income groups that he had defined: under

$25,000; $25,000-$49,999; $50,000-$74,999; and $75,000 or more. For the first three

income groups that are for people with income under $75,000, there were positive gains

on sales with the PromoFlor campaign. Households with incomes of $75,000 or more did

not respond to the PromoFlor programs. Ward estimated that for every dollar spent on

generic advertising at the handler level there was a gain of $5.6 dollars in additional

revenue. Ward concluded that PromoFlor increased by about 10 percent the number of

households buying fresh cut-flowers in a typical month.









Rimal (1998) analyzed the effects of generic and brand promotions of fresh cut-

flowers in the use of retail flower outlets. Rimal used data on household purchases of

fresh cut-flowers through all types of outlets to conduct an expenditure allocation model

that used the Almost Ideal Demand Systems (AIDS) specification to estimate change in

household expenditures on fresh cut-flowers among three flower outlets: florists,

supermarkets and other outlets. Rimal grouped factors that affect expenditure allocation

among outlets in five categories: price effects, expenditure effects, advertising effects,

seasonality effects, and behavioral effects. Rimal found that relative prices, income,

seasonality, attitude and promotion affected expenditure allocation on outlets for fresh

cut-flowers. Rimal concluded that PromoFlor was generally market share neutral;

however, it had significant and positive impacts on florists in the income group $25,000-

$49,999. Overall the impact of brand advertising on florists was positive, while for

supermarkets it was negative.

Girapunthong (2002) analyzed the demand drivers for fresh cut-flowers and their

substitutes. In order to measure the factors influencing the demand for flowers such as

prices, seasonality, and demographic characteristics, she estimated the demand for

flowers in different forms, applying the Almost Ideal Demand Systems (AIDS) to

examine household behaviors in the U.S. flower industry. Her model for fresh cut

flowers, potted and dry/artificial flowers accounted for differences in outlets, purposes,

purchasing occasions, growth, income, demographic characteristics of the households,

and prices. Girapunthong found that all direct price effect coefficients with the seasonal

and actual variables were statistically significant different from zero at the 95 percent

confidence interval. Changes in the relative prices had a significant impact for flower









market shares among fresh cut-flowers, potted flowering plants and dry/artificial flowers.

The coefficient estimates for expenditure effects were significant, indicating that changes

in total household expenditures for flowers had a significant effect on market shares.

Ward (2004) evaluated the impact of a promotion program developed by the

Flower Promotion Organization (FPO) that had the objective of increasing the frequency

of buying of fresh cut flowers among existing female flower buyers in non-traditional

holiday and non-event periods. To measure the impact of the FPO program, Ward

estimated market penetration models and buyer frequency models to determine if the

promotions had stimulated the demand for fresh cut-flowers and then determined the

value of the gains attributed to the promotion program. Ward concluded that the

promotions have impacted the demand for flowers through increasing buyer frequency

and through attracting new buyers. He found that about 87 percent of the increases in

demand for the promotion programs are from the increased transactions per buyer. Ward

found that the demographic group that responded the most to the promotion program was

female buyers that purchases flowers for self-use. This was consistent with the target of

the FPO promotion program.

All of the studies described above have made significant contributions to the

flower industry in the United States. From analyzing attitudes of consumer behavior

toward floral products to estimating the demand for flowers and analyzing the effects of

both generic and brand advertisement programs, these studies have used both mail survey

instruments and extensive household data in order to generator analyses. However, none

of these studies have separated the demand for flowers in terms of market penetration

models and buyer frequency models. Therefore, this research is expected to make






42


substantial contributions in the field of economics and marketing by addressing this

important topic.














CHAPTER 3
UNITED STATES FLORAL INDUSTRY

This chapter consists of descriptive statistics for the flower industry in the United

States. The chapter is organized in six sections. First, an introduction with specifications

on general characteristics for the flower industry and the categorization of the number of

households on each of the divided regions will be presented. Second, expenditures by

region, flower type, purpose of the purchase, gender, age, income and seasonal effects are

presented and discussed. Third, number of transactions by flower type, purpose of the

purchase, gender, age and income are addressed. Fourth, expenditures per transaction by

flower type and demographics are discussed. Fifth, expenditures per buyer or average

price paid by buyers are computed by different demographic characteristics and by flower

type. Sixth, market penetration by flower type, region, purpose of the purchase, age,

gender and income are presented and discussed.

Introduction

The data set for flower purchases from July 1992 to July 2004 was obtained from

the American Floral Endowment (AFE) and Ipsos-NPD group. These data were based on

a consumer panel of several thousand households who reported their purchases of floral

products in the US. The data set is organized by number of households, expenditures,

transactions, and buyers. Each one of these categories contains data on demographic

characteristics including region, purpose of the purchase, product form or flower type,

gender, age and income. There are nine regions throughout this chapter analyzed in terms

of differences on regions based on consumer behavior and demographic characteristics of











the population within each region. The nine regions are New England, Middle Atlantic,


East North Central, West North Central, South Atlantic, East South Central, West South


Central, Mountain and Pacific. There are four basic product forms that are of particular


interest on this study: fresh cut-flowers, flowering and green house, dried or artificial


flowers, and out-door plants. Cut-flowers are subdivided into flower arrangements and


non-arrangements. The purpose of the purchase is either for self-use or to use as gifts.


Gender is either male or female. Age consisted of four categories as follows: 25 years old


or less, 25 to 39 years old, 40 to 54 years old, and 55 years old or more. Income is also


divided into four categories as follows: less than $25,000, $25,000 $49,999, $50,000-


74,999, and $75,000 or more.




Millions

100 Number of Households
Pacific
-- New England 14 Mountain
5 3%
80 Middle Atlantic 31 57ou
146% 5 West
South Central
60 10 9%
East
South Central
62%
40 East
North Central /South Atlantic
16 9% West 18 3%
20 North Central
72%


1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003



Figure 3.1 Total Number of Households by Region from 1993 to 2003. Source: AFE and
Ipsos-NPD group.

The number of households increased 13.16 percent from 73.7 million in 1993 to


83.4 million in 2003. Regions with the greater number of households are South Atlantic,


East North Central, Pacific and Middle Atlantic for a combined value of 64.7 percent of











the total number of households. Figure 3.1 represents the total number of households


from 1993-2003 by region and the aggregate level.


Expenditures For Flowers

Expenditures for flowers increased 82.16 percent from 1993 to 2003. Of the total


expenditures on flowers in the U.S., 18.3 percent belongs to the South Atlantic region,


17.7 percent to East North Central and 16.1 to Middle Atlantic. These three regions


combined represent more than one half the total expenditures in the country. Figure 3.2 is


a graphical representation of the expenditures in the U.S. by regions.



Millions

7 -, I New England
62% acfic
\ 147%
6 Middle Atlantic 14
16 1% Mountain


96%
3 East
North Central -
177% East
2 South Central
5 9%
West South Atlantic
1 North Central 183%
67%
0
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003



Figure 3.2 Total Household Expenditures on Flowers by Region from 1993 to 2003.
Source: AFE and Ipsos-NPD group.

During the 1993 to 2003 period the product form with the highest expenditures


were cut-flowers with a share of 36.3 percent, followed by outdoor with 34.8 percent of


the total expenditures. Figure 3.3 shows the distribution of expenditures by flower type


during the 1993 to 2003 period. Furthermore, fresh cut-flowers were subdivided into


flower arrangements and non-arrangements. Most expenditure of cut-flowers was for


flower arrangements with 54.5 percent of the total expenditures on cut-flowers, as shown


in Figure 3.4.




























Expenditures i Millhons

S00 Dry & Artificial Green House
SOutdoor MCutFlowers

250

200

150

100

050

00
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


Figure 3.3 Total Household Expenditures on Flowers by Flower

2003. Source: AFE and Ipsos-NPD group.


Type, from 1993 to


Green House
197%


Dry & Artificial
91%


Cut Flowers
36 3%


Outdoor
348%


Figure 3.4 Total Household Expenditures on Flowers by flower type, Share of Cut-

Flowers Expenditures from 1993 to 2003. Source: AFE and Ipsos-NPD group.


1993-2003

Outdoor
34 8%


Cut Flowers
36 3%
















Cut Flowers
38 6%


Green House
19 7%


Dry & Artificial
9 1%



2003

Outdoor
367%








Dry & Artificial
54%


Green House
19 3%








47



The purpose of the purchase of floral products was either for self-use or to use as


gifts. During the 1993 to 2003 period, consumers spent about the same amount for self-


use and for gifts including all flower types, all regions and all households. About one half


of the purchases were for self-use and one half for gifts. This relationship is represented


graphically in Figure 3.5.


1993-2003





Expenditures m Milhons 2 003

350 4
300
250
200
2003
150
100
Gift
050 492%
0 00
1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002
Self
508%





Figure 3.5 Total Household Expenditures on Flowers by Purpose from 1993 to 2003.
Source AFE and Ipsos-NPD group.


In terms of gender, females are the group with the largest share of expenditures


with 77.3 percent from the 1993 to 2003 period. Figure 3.6 represents total expenditures


by gender from 1993 to 2003. There is a direct relationship between the age of the


households and the expenditures; the younger the household category, the less it spends


on flowers. This relationship is shown in Figure 3.7. The age group with the largest share


of household expenditures is the oldest age group (55 years of age or more), with 40.9


percent of total household expenditures.

























Expenditures m Millons



MMale
mFemale


2003


1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


Figure 3.6 Total Household Expenditures on

Source: AFE and Ipsos-NPD group.


Flowers by Gender from 1993 to 2003.


1993-2003


Under 25
4 4%


Expenditures m Million


250

200

150

100

0 50

0 00
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


Figure 3.7 Total Household Expenditures on Flowers by Age Groups from 1993 to 2003.

Source: AFE and Ipsos-NPD group.


1993-2003


600


500


400


300 -


200


100


Under 25
35%


2003












The income group with the largest share of expenditures is the highest income


group ($75,000 or more), with 41 percent of the household expenditures. The second


income group ($25,000 to $49,999) was ranked second in terms of household


expenditures with a share of 24 percent of the household's expenditures. The lowest


income group spends the least on flowers, followed by the third income group. These


results suggest that the higher the income the more expenditure on flowers, except for the


third income group ($50,000 to $74,000) where we observe a truncation in expenditure


patterns. This behavior is summarized in Figure 3.8.





1993-2003

Under $25,000 Over $75,000
168% 41 0%



Expenditures in Millions
350
1Under $25,000 1$50,000-$74,999
$00 25,000-$49,999 Over $75,000 $25000-$49 999
240%
2 50 $50,000-$74,999
18 1%
2000

SUndr $25,000
1 00 16 Over $75,000
478%
050
000
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 $25,000-$49,999
215%


$50,000-$74,999
181%




Figure 3.8 Total Household Expenditures on Flowers by Income Groups from 1993 to
2003. Source: AFE and Ipsos-NPD group.


Figure 3.9 shows seasonal pattern expenditures on flowers by flower type. As


shown in the graph, for cut-flowers there is an increase in expenditures during February,


and May, which correspond to Valentine's Day and Mother's Day. Greenhouse, and dry


and artificial flower expenditures increased during the April-May period and December,







50



which corresponds to Mother's Day and Christmas. Outdoor plants have a substantial


increase during the months of April, May and June with a peak of 37 percent of the


expenditures on outdoor plants that occur in May.

Share of Expenditures
040



j-Cut Flowers -Green House
0 30 -Dry/Artificial -Outdoor





020





0 10





000
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Cut Flower 007 0 14 0 08 009 0 13 007 007 006 007 007 007 008
GreenHouse 005 007 008 013 015 007 005 005 005 005 006 017
Dry/Artlficial 007 009 0 10 010 0 12 007 007 007 008 0 07 008 009
outdoor 001 002 007 018 037 014 005 003 005 004 002 001



Figure 3.9 Shares of Seasonal Expenditures on Flowers by Flower Type during the 1993
to 2003 Period. Source: AFE and Ipsos-NPD group.

Transactions on Flowers

The number of transactions is reported on a monthly basis. Differing from regular


consumer demand studies that employ the total quantity consumed on a given period, this


study uses the number of transactions per month as a substitute for the quantity. The


reason for using the number of transactions instead of the total quantity consumed on


flowers is simple. For a given period, a consumer may buy a flower arrangement


consisting of a dozen flowers for example, or that consumer may only buy one flower.


There is no way for the researcher to know the exact quantity of flowers used on different








51



flower products. However, the data utilized in this study are aggregate data of consumer


expenditures for flowers, and the illustration of a single household behavior is used for


exemplification purposes. An alternative approach is to substitute the transactions for the


quantities. The number of transactions by month and by flower type during the 1993 to


2003 period is presented in Figure 3.10.

Number of Transactions
380

3 70 -- Cut Flowers preen House -- - -

3 60

3 50
7pry/Artificial -7 1 1 S






330 I -

320

3 10

3 00

2 90

2- -80-------- -------- ---- -

270 ---- --- -- --- ---- -
3260







2 50
3 0 0 -- -- -- -- --






1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Cut Flowers 273 279 271 285 273 277 279 302 279 279 283
Green House 260 271 258 267 263 278 272 266 265 269 265
Dry/Artificial 316 321 342 356 348 352 361 354 339 346 315
Outdoor 334 336 348 350 341 360 356 376 367 371 359



Figure 3.10 Number of Transactions by Month and by Flower Type During the 1993 to
2003 Period. Source: AFE and Ipsos-NPD group.


The number of transactions is higher for consumers who bought flower products


for self-use, as shown in Figure 3.11. This indicates that consumers who buy floral


products for themselves do it more frequently over a given period. The number of


transactions per buyer over a monthly period has been relatively stable with low


fluctuations from 1993 to 2003.


















1993-2003


Number of Transactions

250 N ift mSef

2 00

1 50

1 00

050

000
1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


2003


Self
582%


Figure 3.11 Number of Transactions by Purpose During the 1993 to 2003 period. Source:

AFE and Ipsos-NPD group.






1 993-2003


Number of Transactions

200/ mMale Female


1 50


1 00


0 50


0 00
1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


Figure 3.12 Number of Transactions by Gender During the 1993 to 2003 period. Source:

AFE and Ipsos-NPD group.












This result is expected, since the majority of consumers who buy product for gifts do so


mostly on calendar occasions. In terms of gender, women complete more transactions


during a given period as compared to men. Figure 3.12 represents the number of


transactions by gender. The number of transactions by age, for the three higher age


groups, has not changed considerably from 1993 to 2003. The only group with high


fluctuations on the number of transactions is the younger group. Figure 3.13 is a


graphical representation of the number of transactions by age. At this point it is not clear


what happened in 1995among the transactions for the youngest age group, households


under 25 years of age.

Number of Transactions
2 00

1 90 4
-under 25 yrs -25-39 yrs
1-40-55 yrs -over 55 yrs I
1ud 25 y 1 18

1 70

1 60

1 50 G u

1 40

1 30 r r

1 20


1 00 -

1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
under25yr 182 155 100 145 181 164 195 156 174 158 181
25-39 y 159 149 160 163 164 170 157 166 170 170 169
40-55 yr 158 168 155 167 165 173 176 172 166 168 168
over55yr 156 160 163 166 159 159 164 171 166 164 172




Figure 3.13 Number of transactions by Age Groups for the 1993 to 2003 Period. Source:
AFE and Ipsos-NPD Group.


The income group with the largest number of transactions is the highest income


group ($75,000 or more). As shown in Figure 3.14, there has been an increase in the











number of transactions for all four-income groups from 1993 to 2003. The lowest income


group, consisting of households with annual income of $25,000 or less, had an increase


of about 12 percent in the number of transactions to 1.69 in 2003, up from 1.51 in 1993.


The number of transactions by income for 2003 is very similar for the four income


groups, differing with less than 3.6 percent from the lowest number of transactions


(second income group) to the highest number of transactions (highest income group).

Number of Transactions
1 85

-Under $25,000 -$50,000-$74,999
1 8 -$25,000-$49,999 -Over $75,000

180 75 I- ---- --- -- --- -----------
1 75


1 70


1 65


1 60 -- -


1 55


1 50
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Under $25,000 151 156 164 164 170 166 163 166 161 168 169
$50,000-$74,999 157 151 154 155 160 170 166 170 171 162 167
$25,000-$49,999 163 162 152 168 153 158 169 165 161 170 172
Over $75,000 164 171 166 177 169 171 175 180 174 170 173
Figure 3.14 Number of Transactions by Income Groups for the 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group.


Figure 3.15 represents seasonality of the number of transactions per month on


flowers by flower type. The seasonality for cut-flowers and greenhouses has been


relatively stable. There have not been dramatic fluctuations in any given month as


represented on the relatively flat curves on the graph. For outdoor plants there is a peak


of 1.96 during May, while dry and artificial plants have two peaks in March and August.


The number of transactions by region is very similar for all regions as represented in














Figure 3.16. The regions with the highest number of transactions are East South Central



and West South Central.


Number of Transactions


1 90


1 80


1 70


1 60


1 50


1 40


1 30


1 20

Cut Fiure
GreenHouse
Dry/At ificial
Outdoor

Figure


Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
132 134 137 133 135 137 137 137 137 132 136 139
132 128 1 30 132 131 129 128 125 123 127 129 130
160 167 181 167 173 167 159 181 166 162 167 153
144 155 172 187 196 181 165 157 159 161 158 138

3.15 Seasonality of the Number of Transactions per Month for the 1993 to 2003

Period. Source: AFE and Ipsos-NPD Group.


Number of Transactions


000 -7 1 ,
New England East South Atlantic West Pacific
North Central South Central
Middle Atlantic West East Mountain
North Central South Central

Figure 3.16 Number of Transactions by Region from the 1993 to 2003 Period. Source:

AFE and Ipsos-NPD Group.











Expenditures per Transaction

Expenditures per transaction are the average amounts of money consumers spend


per transaction. Expressed in a different manner is the average price per transaction. This


is of vital importance, since number of transactions and the average price per transaction


are two key variables for measuring the consumer demand for flowers. Among the four


flower types, consumers spend the most per transaction on cut-flowers as shown in


Figure 3.17. Also, consumers spend more per transaction when the purpose of the


purchase was for gifts. This is represented in Figure 3.18.

Expenditures per Transaction $
35 00

-Outdoor -Green House
I-Dry & Artificial -CutFlowers







25 00 ---





2000 -0-0 -




15 00
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Outdoor 1932 21 63 1622 1753 1790 1861 1978 1964 1898 21 62 21 20
Green House 2003 2099 21 00 1950 21 94 21 11 22 79 23 14 21 72 2287 2607
Dry & Artificial 21 01 1897 1946 1987 1940 23 11 2158 23 90 2827 2442 2801
CutFlowers 24 19 2658 27 08 2699 32 72 3206 3257 3262 3058 2933 2989
Figure 3.17 Expenditures per transaction by Flower Type During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.

Figure 3.19 shows expenditures per transaction by gender with males spending


more per transaction than females. In 2003, males spent 32.24 percent more per


transaction than females. The average expenditure per transaction for males in 2003 was


$17.68, compared to $13.37 for females.














1993-2003


Expenditures per Transaction $
MSelf Gift














1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


Figure 3.18 Expenditures per Transaction by Purpose During the 1993 to 2003 Period.

Source: AFE and Ipsos-NPD Group.


1993-2003


Female
45 3%


Expenditures per Transaction $
mFemale MMale


1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


Figure 3.19 Expenditures per Transaction by Gender During the 1993 to 2003 Period.

Source: AFE and Ipsos-NPD Group.


2000


1500


1000


500


000








58



Consumers of 25 years of age or less have the lowest expenditures per transaction


with an average of $9.4 for 2003. The age group with highest expenditures per


transaction is the category of consumers between 40 and 55 years of age with an average


expenditure per transaction of $14.67 for the year 2003. Figure 3.20 represents the


average expenditure per transaction by age groups. In terms of income, the highest


income group ($75,000 or more) spent the most per transaction, followed by the third


income category ($50,000 to $74,999), with average expenditures per transaction of


$16.15 and $14.45, respectively. The lowest income group ($25,000 or less) showed the


least expenditures per transaction with an average of $11.25 in 2003. Figure 3.21 shows


the average expenditure per transaction for the four income groups during the 1993 to


2003 period.



Expenditures per Transaction $
17 00
-under 25 yrs -25-39 yrs
S -40-55 yrs -over 55 yrs
16 00 - - -

15 00 -

14 00 --/ -

13 00 -- -






1200 -1--- ----I---- ---- ---- ----- --- --- ------- -- -- -- -- -- --
12 00 r











1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
under 25 yrs 760 10 37 8 52 854 10 58 1044 11 75 968 12 30 1086 940
25-39yrs 1094 1395 1240 1399 1414 1333 1369 1427 1368 15 36 1403
40-55 yrs 13 43 1344 15 06 14 02 15 21 14 59 14 76 16 03 15 04 14 64 14 67
over 55yrs 1136 1257 1154 1181 1251 15 00 13 71 1224 1326 1383 1306



Figure 3.20 Average Expenditures per Transaction by Age Groups During the 1993 to
2003 Period. Source: AFE and Ipsos-NPD Group.













Expenditures per Transaction $


-Under $25,000 -$25,000-$49,999
-$50,000-$74.999 -Over $75,000


1995 1996 1997 1998 1999 2000 2001
981 1018 1134 1242 1100 1213 1334
1265 1333 1252 1264 1327 1289 1226
15 80 14 44 1704 1737 15 64 1674 1464
1474 1453 1642 1622 1725 1665 1593


Figure 3.21 Average Expenditures per Transaction by Income Groups During the 1993 to

2003 Period. Source: AFE and Ipsos-NPD Group.


Expenditures per Transaction $


New England East South Atlantic West Pacific
North Central South Central
Middle Atlantic West East Mountain
North Central South Central


Figure 3.22 Average Expenditure per Transaction by Region during

Period. Source: AFE and Ipsos-NPD Group.


the 1993 to 2003


2000



1800


Under $25,000
$25,000-$49,999
$50,000-$74,999
Over $75,000


2000












Figure 3.22 is a graphical representation of the average expenditure per region. The


region with highest expenditures per transaction is East South Central, with an average of


$13.70 during the 1993 to 2003 period. The Pacific region is the region with the lowest


average expenditure per transaction with an average of $10.96 per transaction.


Expenditures Per Buyer


Expenditures per buyer are the average amount by each buyer depending on the


number of transactions that specific buyers had during a monthly period. This measure


will help to describe even further consumer attitudes toward floral products. Average


price paid by consumers was obtained by dividing total expenditures by buyers ($/buyer).


The product form with the highest expenditures per buyer is cut flowers with $40.33 in


2003, up from $31.15 in 1993; this represents almost a 30 percent increase during that


time period. Figure 3.23 is the graphical representation of average expenditures per buyer


from 1993 to 2003.

Expenditures per Buyer $
5000
-CutFlowers -Green House
-Dry & Artificial -Outdoor
45 00 + + --


40 00


35 00


3000 ---


25 00


2000
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
CutFlower 3115 3499 3491 36 72 4238 4205 4334 4577 4132 3945 4033
GreenHouse 2490 2693 2614 25 36 2831 2751 2967 2940 2756 2951 33 73
Dry &Artificial 2945 2926 29 70 3010 31 66 3640 35 10 35 51 41 70 34 54 3933
outdoor 3033 3324 2760 2965 3008 3236 3491 3476 3346 3748 3636
Figure 3.23 Average Expenditures by Buyers by Flower Type During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.








61



Figure 3.24 and 3.25 show the average price paid by buyers by purpose and gender,


respectively.


1993-2003


Expenditures per Buyer $
MSelf MGl I
---_|


15 00
10 00

500
0 00
1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


Figure 3.24 Average Expenditures by Buyers by Purpose During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.


1993-2003


Expenditures per Buyer $
IMale MFemale


1993 1995 1997 1999 2001 2003
1994 1996 1998 2000 2002


Figure 3.25 Average Expenditures by Buyers
Source: AFE and Ipsos-NPD Group.


by Gender During 1993 to 2003 Period.







62



The 25 years of age or less age group is the category with the lowest average


expenditures per buyer with an average of $15.82 in 2003, up from $12.64 in 1993,


which represents a 25 percent increase. The age group between 40 and 55 years of age


has the highest average expenditure per buyer, with an average of $23.47 in 2003, up


from $20.54 in 1993. The lowest age category has some changes over time; in contrast


the other three categories have maintained a trend throughout the period with few abrupt


changes as shown in Figure 3.26.

Expenditures per Buyer $
3000
-under 25 yrs -25-39 yrs
-40-55 yrs -over 55 yrs






2000 --------- --- I -



15 00 -- i



1000



500
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
under 25 yrs 1264 16 15 852 1248 1648 15 28 2247 1437 1943 1731 15 82
25-39 yrs 1654 2027 18 70 21 62 2244 21 16 21 11 2261 21 85 2432 2239
40-55 yrs 20 54 21 75 22 30 22 12 23 88 23 90 24 54 25 74 23 70 23 64 23 47
over55yrs 1672 1892 1775 1819 1874 2260 2101 2012 2054 2144 2130



Figure 3.26 Average Expenditures by Buyers by Age Group During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.

In terms of income, households with overall income of $75,000 or more show the


most expenditures per buyer, followed by the $50,000 to $74,999 group with average


expenditures per transaction of $26.64 and $23.39, respectively. Households with income


of $25,000 or less have the lowest expenditures per buyer with an average of $18.19 in







63



2003. Figure 3.27 shows the average expenditure per buyer for the four income groups


during the 1993 to 2003 period. For all income groups, except for the $50,000 to $74,999


income category, there has been an increasing trend in expenditures per buyer from 1993


to 2003, with some variations within the period.

Expenditures per Buyer $
3000 -
-Under $25,000 -$25,000-$49,999
-$50,000-$74,999 -Over $75,000

25 00 I -- -- -
















1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Under $25,000 1345 1578 1506 1614 1853 1941 1764 1922 20 10 1921 1819
$25,000-$49,999 16 56 19 12 18 59 19 38 18 86 19 85 20 52 20 50 19 70 20 66 20 75
$50,000-$74,999 2207 2563 23 17 2236 2560 2642 2458 26 61 2236 23 53 23 39
Over $75,000 2160 2101 2308 2407 2595 2654 2858 27 78 25 72 27 05 26 64



Figure 3.27 Average Expenditures by Buyers by Income Groups During the 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.

The region with the highest average price paid by buyers is the East South Central


with an average of $21.83, followed by the West South Central region with an average of


$20.59. Mountain is the region with the reported lowest expenditures per buyer among all


regions. Differences in expenditures per buyer by region are very small as shown in


Figure 3.28.











Expenditures per Buyer $


3000



2500


2000


1500


1000


500




New England East South Atlantic West Pacific
North Central South Central
Middle Atlantic West East Mountain
North Central South Central



Figure 3.28 Average Expenditures by Buyers by Region During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group.

Market Penetration

Market penetration is defined as the number of buyers divided by the total number


of households, varying between zero and one on the percentage of households that are


buyers. The market penetration measure is vital for this study since one of its main


objectives is to separate the demand due to market penetration from that of frequency of


buying. Flower types with the highest market penetration are outdoor plants and cut-


flowers as shown in Figure 3.29. Market penetration is higher when the purpose of the


purchase is for self-use as represented in Figure 3.30.














Market Penetration %


S-Cut Flowers -Green House
-Dry/Artificial -Outdoor


1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
CutFlowers 019 017 016 018 018 018 017 017 021 022 027
Green House 017 016 016 017 017 016 016 016 017 017 020
Dry/Artificial 008 008 008 009 007 007 006 006 006 005 005
Outdoor 019 019 019 021 021 021 020 020 022 021 025



Figure 3.29 Percent of Market Penetration by Flower Type During 1993 to 2003 Period.

Source: AFE and Ipsos-NPD Group.


Market Penetration %


0 30






025






020
1I


)93 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003


Figure 3.30 Percent of Market Penetration by Purpose During 1993 to 2003 Period.

Source: AFE and Ipsos-NPD Group.


I L - L -J - I -











Market penetration values are higher for females compared to males as shown in


Figure 3.31. Age groups have a direct relationship with market penetration values, where


the higher the age group, the higher the level of market penetration, as shown in Figure


3.32.

Market Penetration %


SOMALE FEMALE


















11 0 11 12 0 011 0 11 11 0


000 -
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003



Figure 3.31 Percentage Market Penetration by Gender During 1993 to 2003 Period.
Source: AFE and Ipsos-NPD Group.

In Figure 3.33 the percentage values for market penetration by income shows


households with $75,000 or more (highest income group) to have the highest market


penetration values. Surprisingly, the third income group (households with income of


$50,000-$74,999) has the lowest market penetration percentage values among the income


groups.


/

/










67



Market Penetration %
070


i -under 25 yrs -25-39 yrs
060 -40-55 yrs -over 55 yrs



050



040 ---- ------ -



030



0 20 -









1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
under25 yrs 006 007 006 007 006 006 006 006 006 007 007
2539yrs 033 030 028 031 028 025 021 021 021 023 027
40-55yrs 031 030 030 034 036 0031 31 032 035 036 042
over55yrs 048 046 044 049 047 048 050 050 057 054 066


Figure 3.32 Percentage of Market Penetration by Age Groups During 1993 to 2003

Period. Source: AFE and Ipsos-NPD Group.


Market Penetration %
070



0 60

-Under $25,000 -$25,000-$49,999

0 50 -$50,000-$74,999 -Over$75,000



040



030



020



0 10 --- -



000
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003
Under $25,000 030 029 027 031 028 026 023 022 021 020 024
$25,000-$49,999 036 033 030 030 033 030 027 027 029 029 033
$50,000-$74,999 019 017 017 019 017 019 017 018 020 021 025
Over $75,000 033 034 034 041 039 038 040 041 049 050 060


Figure 3.33 Percentage of Market Penetration by Income Groups During 1993 to 2003

Period. Source: AFE and Ipsos-NPD Group.







68


Finally, market penetration among regions is similar ranging from 26 percent to 35


percent in the New England region. Note that in relative terms the South shows the


lowest market penetration levels as shown in Figure 3.34.

Market Penetration %


050


New England East South Atlantic West Pacific
North Central South Central
Middle Atlantic West East Mountain
North Central South Central


Figure 3.34 Percentage of Market Penetration Index by Region During 1993 to 2003
Period. Source: AFE and Ipsos-NPD Group.














CHAPTER 4
CONCEPTUAL FRAMEWORK AND THEORETICAL MODELS

In this chapter the conceptual framework to explain consumer demand for flowers

is presented. The main difference of this study from traditional demand analyses is that

this research separates total demand into two components, that attributed to market

penetration and to changes in buyer frequency. In order to present the conceptual

framework and theoretical models, this chapter has five sections. Consumer demand

theory adapted specifically for the flower case is first presented, followed by a discussion

of the nature of the data. This section shows why the Tobit model is the most appropriate

for the analysis. The conceptual foundations of the Tobit model will be derived and

explained, and then a market penetration model is constructed. Finally, a buyer frequency

model is developed.

Consumer Demand Theory for the Case of Flowers

Consumer demand theory explains consumer preferences or consumer satisfaction

obtained from the physical consumption or use of goods and services. This satisfaction or

preference can be measured by the utility function and its properties, presented in an

earlier chapter. The level of satisfaction can be described as a result of the consumption

or use of a bundle of goods and services. Rational consumers purchase the optimal

quantities of goods and services by maximizing the utility function subject to a budget

constraint, as follows:










(4.1) Max u = u(q,,...,qn)

n
Sto: i pqj =m,
j=1

where pj and q, are the price and the quantity of the jth good, respectively, and m is the

total income. It is assumed that the income is equal to the sum of the expenditures on all

goods; hence, there are no savings, or total income is spent. Traditional demand analyses

are based on the consumption of q, the total quantity consumed. As mentioned

previously, the units for the quantity of flowers are not as straightforward as seen in many

commodities. For example, a person may buy an arrangement of flowers, or a single rose.

The two products can yield to that person similar levels of satisfaction (utility), but the

units for the quantity are different. An alternative approach used to avoid this problem

consists of using the number of transactions in a given period instead of the quantity. If

one can obtain the number of buyers and the number of transactions for a given period,

then the units in the model can be expressed as the total frequency of purchasing or

number of buyers times transaction per buyer (Girapunthong, 2002). Let b represent the

number of buyers andf represent the frequency of buying, then the income constraint on

(4.1) can be modified to obtain:

n n
(4.2) ipjfjbj m= ,
J=1 J=1

where pj represents the price per transaction. The Lagrangian procedure can now be

used on the utility function, and then the consumer's demand function becomes:

(4.3) fb, =f(m,p,,pj).









The demand function now describes the consumption of goods in terms of a particular

income, prices of the jth goods in quantity terms and prices of the ith good in transaction

units or dollars per transaction. This would be the utility function maximized in order to

obtain the demand for flowers.

An Overview of NPD Data Set

The data set used in the models was obtained from the Ipsos and NPD group; the

relevant variables used in this study are described on Chapter 3. In order to separate the

total effect of the demand for flowers in the U.S. into market penetration demand and

buyer frequency demand, market penetration and buyer frequency models are developed.

For each model two approaches will be used. First regional and product form changes are

allowed to have different intercepts and slopes. This means that there will be a different

equation for each region and product form. Hence, there will be i x j equations, where i is

the number of product forms and j is the number of regions. The second approach will

allow changes in the intercept and slopes product forms, while regional changes are only

in the slopes. The intercept of the regional changes, as well as the other dummy variables

employed, will represent the average of all regions. This approach will have i equations,

or one equation for each flower product form, with the forms being cut-flowers,

flowering plants, dry/artificial and out-door.

For the first approach, the dependent variables for each of the models will be

penetration and frequency. Penetration is easily calculated by dividing the number of

buyers by the number of households as in equation (4.4):

BUY
(4.4) PEN, -- 0 < PEN, < 1,
HWD 1









where PENn BUY and HWD, are penetration, buyers and households for ith product

form and thejth region.

Frequency is derived by dividing transactions by buyers, as in equation (4.5):


(4.5a) FQ, TR 2AJ
BUY

where FQ, TRN, and BUY are frequency, transactions and buyers for the ith

product form and the jth region. By definition a person who is a buyer had at least one

transaction or more in a given period, or else that person would not be defined as a buyer.

Since FQ is censored at 1, an often-used option for estimation purposes is to adjust the

censored variable so that the lower limit is zero. That adjustment simply entails

subtracting the lower level from the original censored value. That is

(4.5b) FRQ, = FQ, -1,

where FRQJ is the adjusted frequency.

The independent variables for both penetration and frequency models will be

discrete variables created for income, gender, purpose, age and seasonal monthly

dummies. If we employ the common method of creating dummy variables described by

Greene (2000), then the base level for all the coefficients of the dummy variables will be

the category we left out of the equations in order the evade the dummy variable trap. A

different approach consists of restricting the sum of the coefficient of the dummy

variables to zero. In this case, the base of the dummies would be the mean of all the

categories. For example, let 8P, be the parameter estimate for income, then if the

4 4
restriction 0/k, =0 is imposed, then /1 =- ,/k, is obtained and then the dummy
k=1 k=2









variable dinck = inck -inc1 will be created, where k # 1. More generally we would

impose the restriction as follows:

K K
(4.6) = 0 to obtain /, = -Zf k ,
k=l k=2

and in order to create the dummy variables the following operation follows:

(4.7) dummyk = categoryk category, where k # 1.

The price per transaction is calculated from the data set by dividing expenditures by

the transactions:

EXP
(4.8) PRT = j,
D TRNA I

where PRT EXP and TRN, are price per transaction, expenditures and transactions

for the ith product form and the jth region. A problem arises for the penetration model; if

a household was not a consumer by definition the number of transactions for that

household in that period was zero, and then the price would not be defined and in the data

set would be considered zero. This would result in an underestimation of the price, since

the zeros would lower the demand effect for the market penetration model. Therefore

price per transaction will only be included in the frequency models where purchases

occurred.

The second approach is very similar. The only difference is that instead of allowing

the intercept to change for each region, that is calculating an equation for each region,

dummy variables for regions will be created using the process described above; and then

these dummy variables for the different regions will be added as independent variables to

both penetration and frequency models.










When a data set has a dependent variable that is zero for a significant fraction of the

observations, conventional regression methods fail to account for the qualitative

difference between limit (zero) observations and non-limit (continuous) observations

(Greene, 2000). Tobit models have been used in single commodity studies to account for

the fact that not all households purchase that commodity. Greene (2000) showed that in a

regression model where a large proportion of the observations of the dependent variable

are zero, then the ordinary least squares (OLS) parameter estimates tend to be biased

toward zero, and the degree of bias depends on the amount of censoring (Cornick et al.,

1994). For these type of data the most appropriate model is the one developed by Tobin,

the Censored Regression Model or simply the Tobit model.

Tables 4.1 and 4.2 show the percentage of observations for which the dependent

variable for penetration models are zero, and the percentage of the observations for which

the dependent variable for the frequency models are one. Note that the frequency table

show the percentage of households that are in the censored limit of one, since it is more

meaningful to explain in the table as the households who made only one transaction in a

given period, and hence became buyers.

Table 4.1 Percentage of Observations of Penetration Model Dependent Variable That Are
Censored at Zero. Source: AFE and Ipsos Group.

% ofpen=0 ALL NE MA ENC WNC SA ESC WSC MOUNT PACIF
TOTAL 5.87% 44.76% 24.72% 26.07% 45.01% 24.44% 49.64% 36.33% 47.41% 25.63%
indoor 8.08% 51.07% 30.25% 31.30% 51.09% 31.12% 56.42% 43.54% 54.98% 32.52%
cutflowers I .. i ".. 45 15" .. .I ".. JI "..
A range I ,'".. -- .. .. Is .. ." .. 4 ..".. -I, .. -', "*
N onA rran 2'1". ,. s;".. in" ., ,ii* ".. -Iii,"., ',II;" ;; 4 ".. r. i,-"., i,-". I I5 .
GreenH house 1 2"i .. .. 5J I'".. 51 -1".. i'" .. 5 1".. "5 I.. .. .. .. 51 ..
Flower 27.68% 78.90% 63.16% 61.87% 78.84% 61.88% 82.26% 73.88% 81.46% 62.79%
Foliage 33.58% 87.46% 73.69% 70.55% 84.05% 70.21% 87.07% 77.92% 86.44% 68.95%
Dry & Artificial 44.12% 90.99% 76.22% 71.54% 81.39% 70.08% 81.58% 77.16% 87.93% 80.19%
Outdoor 26.95% 76.45% 63.16% 63.94% 76.50% 55.23% 77.47% 68.55% 77.96% 55.87%
TOTAL 5.87% 44.76% 24.72% 26.07% 45.01% 24.44% 49.64% 36.33% 47.41% 25.63%












Table 4.2 Percentage of Observations of Frequency Model Dependent Variable That Are
Censored at One. Source: AFE and Ipsos Group.
% offreq 1 ALL NE MA ENC WNC SA ESC WSC MOUNT PACIF
TOTAL 1 .. 45: 1".. 4 ".. 4'- .. 45 I .. : ".. 4 ". -,41 -I" '4"' -.. -"..
indoor '".. 5 I`" .. 4 14"".. l1 ,.".. 54 41".. I ',".. .5r," .. 4 ".. ., .. 4 5 ,"..
cutflow ers :2. .. .-I -.".I 4 ".. -" ,. '". -" .. N'",. -" .. ". j '"...
A range ".. ...4".. ".. ,".. '".. .. I".. 4".. l"..
NonArran 33.20% 66.10% 58.21% 60.56% 73.85% 61.15% 75.93% 65.31% 73.13% 61.22%
Green House 39.64% 69.05% 60.32% 60.14% 70.83% 60.40% 69.81% 66.01% 71.83% 59.27%
Flower 49.06% 74.91% 66.95% 68.24% 76.87% 68.00% 76.13% 73.20% 78.70% 67.72%
Foliage 55.94% 76.38% 74.89% 69.71% 77.76% 72.35% 77.77% 74.94% 75.28% 69.60%
Dry & Artificial 35.34% 62.05% 54.88% 50.13% 54.29% 51.80% 56.36% 54.20% 64.93% 55.20%
Outdoor 14". 4 .. -4 ., 4' i,'".. ',.l 4'' 4- ", -4 p".. -'"., 4 '. .. A- Ii r.4" ,
TOTAL 1 ".. 4' ".. '4 ".. ; ".. 45 I".. ;: I ".. 4, .. 41 I_".. 4" ; I ..



In our household data set a large proportion of the dependent variables for both models

are censored at zero or one. Therefore, procedures that take into consideration censored

expenditures distributions must be used (i.e., the Tobit model).

The Tobit Model

Nobel prize winner Tobin (1958), showed that when the dependent variable in a

regression model equation has a lower or upper limit, and the dependent variable takes

the value of the limit for a large number of sample observations, conventional multiple

regression analysis is not an appropriate technique to be used (Lung-Fei and Maddala,

1985). As shown in Tables 4.1 and 4.2, a large proportion of the penetration and

frequency data take the value of the limit. In order to account for this truncation on the

data set, Tobin developed a model specified as follows:

(4.9) y,* = xp1 + E,,


where x, is a (1 x K) vector of explanatory variables and E, N(0,o2) and it is


independent of other errors. The problem arises because in order for a household to be a

buyer, it has to have at least one transaction during a given period. Adjusting the









subtracted one from the frequency variable to have the lower limit equal zero. In the

penetration model a large number of the observations take the value of the lower limit,

zero. Thus for any household the penetration and frequency models would take the form:

(4.10) y, = y, if y, > 0

y, =0 if y,' <0.

From the total number of observations T in the sample, the number of observations

can be divided into two groups; one for which y, = 0 TO; and another for the number of

observations for which y, > 0, T,. In order to observe the statistical problems arising

from the censored sample problem, consider leaving out of the analysis the To

observations for which y, =0. For the remaining T, sample observations, they are

complete observations. Hence, one can use least squares estimators to estimate 8. The

problem is that this estimator is biased and inconsistent. In order to prove that, one can

write down the expectation of the observed values of y, conditional on the fact that

y, >O:

(4.11) Ey, > 0] = x:' + E(E, Y, > 0).

If the conditional expectation of the error term is zero, then the estimates of the least

square regression on T, would provide an unbiased estimator for /. However this is not

the case; if the E, are independent and normally distributed random variables, then the

expectation would be:

(4.12) E[E, I y, > 0]= E[E, I E, > -x:'] > 0.

It can be shown that this conditional expectation can also be expressed in the following

manner:









(4.13) E[>, >-x,']= ,


where and (, are the standard normal probability distribution function (p.d.f), and

cumulative distribution function (c.d.f) evaluated at (x,~l/a); therefore in the regression

model, if y, > 0, then,

y, = x'p + ,
(4.14) +
= x'Y +a +u,


if we apply the regular least squares procedures the term is omitted. Since that


term is not independent of x, the results are biased and inconsistent.

In order to estimate the parameters / and 02 consistently, maximum likelihood

estimation (MLE) procedures can be used. The likelihood function of the sample has a

component for the observations that are positive, and one for the observations that are

zero. For the observations y, = 0 it is known that x,' + e, < 0 or expressed in a different

way, E, < -x,', then,


(4.15) Pr[y, = 0]= Pr[E, < -x:/f]= Pr E < -X: = 1-,.


If we define the product of the observations over the zero lower limit level to be TI and

the product over the positive observations to be H1, the likelihood function of the Tobit

model is given by:

(4.16) 2= H,(1- ,n)Hj27r 22 exp{-(y -x:(f)2/22}

The corresponding log-likelihood function would be:









(4.17) L=ln= ln(1-( ,)-(T,/2)ln(2r)-(T,/2)1nC2 E (y f)2

Then the first order conditions are:

OL 1 E0x 1
(4.18a) 1 + Y (y '1)x,


OL 1 (x-'p8) T 1I
(4.18b) 1,C 1 (,) T (y, + 2 I)
20 2-3 1 2-2 20-4E

The Time Series Processor (TSP) has a routine to maximize the log likelihood

function for the Tobit model. The Tobit routine uses the analytic first and second

derivatives to obtain maximum likelihood estimates via the Newton-Raphson algorithm.

The starting values for the parameters are obtained from a regression on the observations

with positive y values. The numerical implementation involves evaluating the normal

density and cumulative normal distribution functions. The cumulative distribution

function is computed from an asymptotic expansion, since it has no closed form. The

ratio of the density to the distribution function, used in the derivatives, is also known as

the Inverse Mills Ratio (Hall, 1992).

Market Penetration Models

There are two approaches used to develop the market penetration models. The first

approach, Penetration Model I, uses a Tobit model equation for each product form i, and

marketing region. In consequence there will be one equation for each product form and

region, for a total of i x j equations for the first approach. This approaches captures

changes in both the intercept and the slope of the independent variables for market

penetration. The second approach, Penetration Model II, incorporates regions as









independent dummy variables. Therefore the regions would have a common slope but

different intercepts depending on the regional dummy estimates.

Penetration Model I

Let X, represent income, gender, purpose, age, and seasonal monthly expenditures.

Then the general Model I would be expressed with any reference to the observation being

dropped for notational convenience as follows:

(4.19) PEN, = PEN, = X, + u, if PEN, > 0

PENj = 0 if PEN, <0,

which is the classical Tobit model described in the previous section. The complete model

one would be represented as follows:


PEN, = 5o0,) + 5k(U) (INCk(U) -NCl,()+ 56(U)(GEN2() GENl )+
k=2

(4.20) 3 8(PUR 2(W) U -PUR,)+ )(AGE( AGEJ)+
k=2
12
112+k(S) (MTHk() -MTH() )
k=2

where the variables are explained in Table 4.3

Penetration Model II

The Penetration Model II is very similar to the market penetration model I, except

that the regions are included in the model as dummy variables. Therefore in this model

we will have fewer equations, one for each product form. Let us again define the Market

Penetration Model II by PEN, which is also a number between zero and one and using

X, as shown in equation (4.19). Now X, includes the additional regional variable added

to equation (4.20). Note that thej subscripts for the regions are now dropped:










PEN =(5(1) +1 Zk(i)((I\Tk(i) INC1())+ 56(IGEA2(,) -GI Nl(,)+
k2

(4.21) (58(,#UR2(i) _P UR 1(1) + 4
k-


312+k(j)(MTHk(z) -MTHI,)+ ,25 k() REGk(,)
k=2 k-2

Table 4.3. Variables for the Market Penetration Model I


INCOME INC2 = ($25,000 $49,999 = 1) or (otherwise = 0)
INC3 = ($50,000 $74,999 = 1) or (otherwise = 0)
INC4 = ($75,000 or more = 1) or (otherwise = 0)
GENDER GEN2 = (male = 0) and (female = 1)
PURPOSE PUR2 = (self = 0) and (gift = 1)
AGE AGE2 = (25 50 = 1) or (otherwise = 0)
AGE3 = (50 75 = 1) or (otherwise = 0)
AGE4 = (75 or more = 1) or (otherwise = 0)
SEASON MTH2 = (February = 1) or (otherwise = 0)
MTH3 = (March = 1) or (otherwise = 0)
MTH4 = (April = 1) or (otherwise = 0)
MTH5 = (May = 1) or (otherwise = 0)
MTH6 = (June = 1) or (otherwise = 0)
MTH7 = (July = 1) or (otherwise = 0)
MTH8 = (August = 1) or (otherwise = 0)
MTH9 = (September = 1) or (otherwise = 0)
MTH10 = (October = 1) or (otherwise = 0)
MTH11 = (November = 1) or (otherwise = 0)
MTH12 = (December = 1) or (otherwise = 0)


where the variables are explained in table 4.4.

Buyer Frequency Models

The frequency models follow the same structure as the market penetration models,

with the two approaches. Therefore the structures are Buyer Frequency Model I with i xj


equations, and Buyer Frequency model II with i equations.


REGI (1) ) + U




Full Text

PAGE 1

MEASURING DEMAND FACTORS INFLUENCING MARKET PENETRATION AND BUYING FREQUENCY FOR FLOWERS By MARCO ANTONIO PALMA GARCIA A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Marco Antonio Palma Garcia

PAGE 3

This document is dedicated to my parents An tonia and Marco, and my siblings Eric and Vanessa, with gratitude and love.

PAGE 4

iv ACKNOWLEDGMENTS I will begin by thanking the Almighty for giving me life, and this educational opportunity. This work would not have been possible without the continuous support, guidance, patience and time dedicated from my committee chair, Dr. Ronald W. Ward. I also want to extend my most sincere appreci ation to the rest of my committee members for their support: Dr. Lisa House, Dr. Ma rk Brown, Dr. Tom Sheehan and Dr. Gerhard Schiefer. I would like to thank my family, speciall y Antonia, Marco, Vanessa, Roy, Carolina, Eric Sebastian, Ivette, Alicia, Doris and Moni ca, for their affection, love, encouragement and support always, specially during difficu lt times. A special thank you goes to my friends Fausto, Ricky, Jorge, Angel, Sergio, Claudio, and my friends in Gainesville, for being a source of strength and motivati on during my graduate school experience. Finally I would like to thank my classmat es Shiferaw, Mariana, Angel, Lurleen, Ronald, Joy, and Raphael for their continuous friendship, and support during all of my graduate school experience, especially during the core of the program. I also want to thank the faculty and staff of the Food and Resource Economics Department of the University of Florida for their assistance, especially Dr. Charles Adams, Dr. Richard Beilock, Dr. Tom Spreen, Dr. William Messina, Dr. Jane Luzar, Dr. Jeff Burkhardt, Brian Sevier and Jessica Herman.

PAGE 5

v TABLE OF CONTENTS page ACKNOWLEDGMENTS.................................................................................................iv LIST OF TABLES...........................................................................................................viii LIST OF FIGURES.........................................................................................................xiv ABSTRACT.....................................................................................................................xi x CHAPTER 1 INTRODUCTION........................................................................................................1 Overview of the Industry..............................................................................................3 Problem Statement......................................................................................................10 Objectives...................................................................................................................11 Research Methodology...............................................................................................11 Data and Scope...........................................................................................................13 Organization of the Study...........................................................................................14 2 LITERATURE REVIEW...........................................................................................15 Consumer Behavior....................................................................................................15 Consumer Demand Analysis......................................................................................20 Utility Maximization and Demand Functions.....................................................20 Properties of the Demand Functions...................................................................26 Adding-up restriction...................................................................................26 Symmetry restriction....................................................................................27 Negativity restriction....................................................................................27 Homogeneity restriction...............................................................................28 Marketing Research Models.......................................................................................28 Market Penetration Models.................................................................................29 Logistic function..........................................................................................29 The Pyatt function........................................................................................30 The Gompertz function................................................................................31 The Weblus model.......................................................................................31

PAGE 6

vi The log-inverse function..............................................................................32 The lognormal function................................................................................33 Repeat Buying Models........................................................................................33 Review of Past Studies on Flower Products...............................................................36 3 UNITED STATES FLORAL INDUSTRY................................................................43 Introduction.................................................................................................................43 Expenditures on Flowers............................................................................................45 Transactions on Flowers.............................................................................................50 Expenditures per Transaction.....................................................................................56 Expenditures Per Buyer..............................................................................................60 Market Penetration......................................................................................................64 4 CONCEPTUAL FRAMEWORK AND THEORETICAL MODELS.......................69 Consumer Demand Theory for the Case of Flowers..................................................69 An Overview of NPD Data Set...................................................................................71 The Tobit Model.........................................................................................................75 Market Penetration Models.........................................................................................78 Penetration Model I.............................................................................................79 Penetration Model II............................................................................................79 Buyer Frequency Models............................................................................................80 Frequency Model I..............................................................................................81 Frequency Model II.............................................................................................82 5 EMPIRICAL RESULTS............................................................................................84 Data Usage..................................................................................................................84 Demand Model Equations..........................................................................................85 Model I Results...........................................................................................................87 Market Penetration Model Results I....................................................................87 Buyer Frequency Model Results I.......................................................................93 Model II Results.........................................................................................................99 Market Penetration Model Results II................................................................100 Buyer Frequency Model Results II....................................................................109 6 SIMULATION ANALYSIS.....................................................................................119 Introduction...............................................................................................................119 Simulations For Cut-Flowers....................................................................................123 Simulations For Flowering Plants And Greens........................................................131 Simulations For Dry/Ar tificial Flowers....................................................................139 Simulations For Outdoor..........................................................................................147 7 SUMMARY AND CONCLUSIONS.......................................................................157

PAGE 7

vii Summary and Conclusions.......................................................................................157 Limitations and Directio n for Future Research........................................................167 APPENDIX A MODEL I RESULTS...............................................................................................169 B TSP PROGRAMS....................................................................................................241 LIST OF REFERENCES.................................................................................................256 BIOGRAPHICAL SKETCH...........................................................................................259

PAGE 8

viii LIST OF TABLES Table page 4.1 Percentage of Observations of Pene tration Model Dependent Variable That Are Censored at Zero. Source: AFE and Ipsos Group.............................................74 4.2 Percentage of Observations of Freque ncy Model Dependent Variable That Are Censored at One. Source: AFE and Ipsos Group.....................................................75 4.3 Variables for the Market Penetration Model I.........................................................80 4.4 Variables for the Market Penetration Model II........................................................81 5.1 Distribution of the States for Each Region...............................................................86 5.2 General Statistical Information About the Market Penetration Model by Flower Type..........................................................................................................................8 8 5.3 Market Penetration Parameter Estimates and T-Values for Indoor and CutFlowers.....................................................................................................................89 5.4 Market Penetration Parameter Estimates and T-Values for Flower Arrangements and Non-Arrangements.....................................................................90 5.5 Market Penetration Parameter Estim ates and T-Values for Plants and Dry/Artificial Flowers..............................................................................................91 5.6 Market Penetration Parameter Estimates and T-Values for Outdoor Flowers.........92 5.7 General Statistical Information Abou t the Buyer Frequency Model by Flower Type..........................................................................................................................9 5 5.8 Buyer Frequency Parameter Estimates and T-values for Indoor and CutFlowers.....................................................................................................................96 5.9 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements and Non-Arrangements............................................................................................96 5.10 Buyer Frequency Parameter Estimates and T-values for Plants and Dry/Artificial............................................................................................................97 5.11 Buyer Frequency Parameter Estimates and T-values for Outdoor...........................97

PAGE 9

ix 5.12 General Statistical Information About the Market Penetration Model II by Flower Type...........................................................................................................100 5.13 Market Penetration Parameter Estimates and T-Values for Indoor and CutFlowers...................................................................................................................101 5.14 Market Penetration Parameter Estimates and T-Values for Flower Arrangements and Non-Arrangements...................................................................102 5.15 Market Penetration Parameter Estim ates and T-Values for Plants and Dry/Artificial Flowers............................................................................................103 5.16 Market Penetration Parameter Estimates and T-Values for Outdoor Flowers.......104 5.17 General Statistical Information Ab out the Buyer Frequency Model II by Flower Type...........................................................................................................109 5.18 Buyer Frequency Parameter Estimates and T-values for Indoor and CutFlowers...................................................................................................................110 5.19 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements and Non-Arrangements..........................................................................................111 5.20 Buyer Frequency Parameter Estimates and T-values for Plants and Dry/Artificial..........................................................................................................112 5.21 Buyer Frequency Parameter Estimates and T-values for Outdoor.........................113 A.1 Market Penetration Model I Results for Indoor and Cut-Flowers in New England...................................................................................................................169 A.2 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in New England..............................................................................170 A.3 Market Penetration Model I Results for Plants and Dry/Artificial in New England...................................................................................................................171 A.4 Market Penetration Model I Results for Outdoor in New England........................172 A.5 Buyer Frequency Model I Results for Indoor and Cut-Flowers in New England..173 A.6 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in New England..............................................................................174 A.7 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in New England...................................................................................................................175 A.8 Buyer Frequency Model I Results for Outdoor in New England...........................176

PAGE 10

x A.9 Market Penetration Model I Results for Indoor and Cut-Flowers in Middle Atlantic...................................................................................................................177 A.10 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in Middle Atlantic...........................................................................178 A.11 Market Penetration Model I Results fo r Plants and Dry/Ar tificial in Middle Atlantic...................................................................................................................179 A.12 Market Penetration Model I Resu lts for Outdoor in Middle Atlantic....................180 A.13 Buyer Frequency Model I Results for Indoor and Cut-Flowers in Middle Atlantic...................................................................................................................181 A.14 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in Middle Atlantic...........................................................................182 A.15 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in Middle Atlantic...................................................................................................................183 A.16 Buyer Frequency Model I Results for Outdoor in Middle Atlantic.......................184 A.17 Market Penetration Model I Results fo r Indoor and Cut-Flowers in East North Central....................................................................................................................185 A.18 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in East North Central......................................................................186 A.19 Market Penetration Model I Results for Plants and Dry/Artific ial in East North Central....................................................................................................................187 A.20 Market Penetration Model I Results for Outdoor in East North Central...............188 A.21 Buyer Frequency Model I Results for Indoor and Cut-Flowers in East North Central....................................................................................................................189 A.22 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in East North Central......................................................................190 A.23 Buyer Frequency Model I Results for Pl ants and Dry/Artificial in East North Central....................................................................................................................191 A.24 Buyer Frequency Model I Results fo r Outdoor in East North Central...................192 A.25 Market Penetration Model I Results fo r Indoor and Cut-Flowers in West North Central....................................................................................................................193 A.26 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in West North Central.....................................................................194

PAGE 11

xi A.27 Market Penetration Model I Results for Plants and Dry/Artificial in West North Central....................................................................................................................195 A.28 Market Penetration Model I Results for Outdoor in West North Central..............196 A.29 Buyer Frequency Model I Results for Indoor and Cut-Flower s in West North Central....................................................................................................................197 A.30 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in West North Central.....................................................................198 A.31 Buyer Frequency Model I Results for Pl ants and Dry/Artificial in West North Central....................................................................................................................199 A.32 Buyer Frequency Model I Results fo r Outdoor in West North Central.................200 A.33 Market Penetration Model I Results for Indoor and Cut-Flowers in South Atlantic...................................................................................................................201 A.34 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in South Atlantic.............................................................................202 A.35 Market Penetration Model I Results for Plants and Dry/Artificial in South Atlantic...................................................................................................................203 A.36 Market Penetration Model I Resu lts for Outdoor in South Atlantic......................204 A.37 Buyer Frequency Model I Results for Indoor and Cut-Flowers in South Atlantic...................................................................................................................205 A.38 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in South Atlantic.............................................................................206 A.39 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in South Atlantic...................................................................................................................207 A.40 Buyer Frequency Model I Results for Outdoor in South Atlantic.........................208 A.41 Market Penetration Model I Results fo r Indoor and Cut-Flowers in East South Central....................................................................................................................209 A.42 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in East South Central......................................................................210 A.43 Market Penetration Model I Results for Plants and Dry/Artific ial in East South Central....................................................................................................................211 A.44 Market Penetration Model I Results for Outdoor in East South Central...............212

PAGE 12

xii A.45 Buyer Frequency Model I Results for Indoor and Cut-Flowers in East South Central....................................................................................................................213 A.46 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in East South Central......................................................................214 A.47 Buyer Frequency Model I Results for Pl ants and Dry/Artificial in East South Central....................................................................................................................215 A.48 Buyer Frequency Model I Results fo r Outdoor in East South Central...................216 A.49 Market Penetration Model I Results fo r Indoor and Cut-Flowers in West South Central....................................................................................................................217 A.50 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in West South Central.....................................................................218 A.51 Market Penetration Model I Results for Plants and Dry/Artificial in West South Central....................................................................................................................219 A.52 Market Penetration Model I Results for Outdoor in West South Central..............220 A.53 Buyer Frequency Model I Results for Indoor and Cut-Flower s in West South Central....................................................................................................................221 A.54 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in West South Central.....................................................................222 A.55 Buyer Frequency Model I Results for Pl ants and Dry/Artificial in West South Central....................................................................................................................223 A.56 Buyer Frequency Model I Results fo r Outdoor in West South Central.................224 A.57 Market Penetration Model I Results for Indoor and Cut-Flowers in Mountain.....225 A.58 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in Mountain.....................................................................................226 A.59 Market Penetration Model I Results fo r Plants and Dry/Artif icial in Mountain....227 A.60 Market Penetration Model I Results for Outdoor in Mountain..............................228 A.61 Buyer Frequency Model I Results fo r Indoor and Cut-Flowers in Mountain........229 A.62 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in Mountain.....................................................................................230 A.63 Buyer Frequency Model I Results for Plants and Dry/Artif icial in Mountain.......231

PAGE 13

xiii A.64 Buyer Frequency Model I Resu lts for Outdoor in Mountain.................................232 A.65 Market Penetration Model I Results for Indoor and Cut-Flowers in Pacific.........233 A.66 Market Penetration Model I Resu lts for Flower Arrangements and NonArrangements in Pacific.........................................................................................234 A.67 Market Penetration Model I Results fo r Plants and Dry/Artificial in Pacific........235 A.68 Market Penetration Model I Results for Outdoor in Pacific..................................236 A.69 Buyer Frequency Model I Results fo r Indoor and Cut-Flow ers in Pacific............237 A.70 Buyer Frequency Model I Results for Flower Arrangements and NonArrangements in Pacific.........................................................................................238 A.71 Buyer Frequency Model I Results for Plants and Dry/Artificial in Pacific...........239 A.72 Buyer Frequency Model I Resu lts for Outdoor in Pacific.....................................240

PAGE 14

xiv LIST OF FIGURES Figure page 1.1 Large Grower Sales by Crop Group From 1985 to 2003. Source. Economic Research Service, USDA, 2003.................................................................................5 1.2 Total Number of Large Growers in the US From 1992 to 2003. Source. Economic Research Service, USDA, 2003................................................................5 1.3 Average Sales Per Large Grower by Crop Group from 1992 to 2003. Source. Economic Research Service, USDA, 2003................................................................6 1.4 US Total Imports and Exports from 1976 to 2003. Source. Economic Research Service, USDA, 2003.................................................................................................6 1.5 US Imports of Cut Flowers and Nu rsery Crops by Country, 2003. Source. Economic Research Service, USDA, 2003................................................................7 1.6 US Exports of Cut Flowers and Nu rsery Crops by Country, 2003. Source. Economic Research Service, USDA, 2003................................................................7 2.1 Graphic Presentation of Indifference Curves...........................................................18 3.1 Total Number of Households by Re gion from 1993 to 2003. Source: AFE and Ipsos-NPD group......................................................................................................44 3.2 Total Household Expenditures on Flowers by Region from 1993 to 2003. Source: AFE and Ipsos-NPD group.........................................................................45 3.3 Total Household Expenditures on Flowers by Flower Type, from 1993 to 2003. Source: AFE and Ipsos-NPD group...............................................................46 3.4 Total Household Expenditures on Flow ers by flower type, Share of CutFlowers Expenditures from 1993 to 2003. Source: AFE and Ipsos-NPD group.....46 3.5 Total Household Expenditures on Flowers by Purpose from 1993 to 2003. Source AFE and Ipsos-NPD group..........................................................................47 3.6 Total Household Expenditures on Flowers by Gender from 1993 to 2003. Source: AFE and Ipsos-NPD group.........................................................................48

PAGE 15

xv 3.7 Total Household Expenditures on Flowers by Age Groups from 1993 to 2003. Source: AFE and Ipsos-NPD group.........................................................................48 3.8 Total Household Expenditures on Flowers by Income Groups from 1993 to 2003. Source: AFE and Ipsos-NPD group...............................................................49 3.9 Shares of Seasonal Expenditures on Flowers by Flower Type during the 1993 to 2003 Period. Source: AFE and Ipsos-NPD group................................................50 3.10 Number of Transactions by Month and by Flower Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD group....................................................51 3.11 Number of Transactions by Purpos e During the 1993 to 2003 period. Source: AFE and Ipsos-NPD group......................................................................................52 3.12 Number of Transactions by Gender During the 1993 to 2003 period. Source: AFE and Ipsos-NPD group......................................................................................52 3.13 Number of transactions by Age Gr oups for the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group......................................................................................53 3.14 Number of Transactions by Inco me Groups for the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................54 3.15 Seasonality of the Number of Tran sactions per Month for the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................55 3.16 Number of Transactions by Regi on from the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group......................................................................................55 3.17 Expenditures per transaction by Flower Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................56 3.18 Expenditures per Transaction by Pu rpose During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................57 3.19 Expenditures per Transaction by Gender During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................57 3.20 Average Expenditures per Transac tion by Age Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group...................................................58 3.21 Average Expenditures per Transacti on by Income Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group...................................................59 3.22 Average Expenditure per Transac tion by Region during the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................59

PAGE 16

xvi 3.23 Average Expenditures by Buyers by Flower Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................60 3.24 Average Expenditures by Buyers by Purpose During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................61 3.25 Average Expenditures by Buyers by Gender During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................61 3.26 Average Expenditures by Buyers by Age Group During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................62 3.27 Average Expenditures by Buyers by Income Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................63 3.28 Average Expenditures by Buyers by Region During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................64 3.29 Percent of Market Penetration by Flower Type During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................65 3.30 Percent of Market Penetration by Purpose During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group......................................................................................65 3.31 Percentage Market Penetration by Gender During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group......................................................................................66 3.32 Percentage of Market Penetration by Age Groups During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group........................................................................67 3.33 Percentage of Market Penetrati on by Income Groups During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................67 3.34 Percentage of Market Penetra tion Index by Region During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group............................................................68 5.1 Flower Type Groups.................................................................................................85 5.2 Market Penetration Seasonality for Cut-Flowers and Plants...................................93 5.3 Market Penetration Seasonality for Dry/Artificial and Outdoor flowers.................94 5.4 Buyer Frequency Seasonality for Cut-Flowers and Plants.......................................98 5.5 Buyer Frequency Seasonality for Dr y/Artificial and Outdoor Flowers...................99 5.6 Market Penetration Seasonality for Cut-Flowers and Plants.................................105 5.7 Market Penetration Seasonality for Dry/Artificial and Outdoor flowers...............106

PAGE 17

xvii 5.8 Market Penetration Regional Cha nges for Cut-Flowers and Plants.......................107 5.9 Market Penetration Regional Changes fo r Dry/Artificial and Outdoor Flowers...108 5.10 Buyer Frequency Seasonality for Cut-Flowers and Plants.....................................114 5.11 Buyer Frequency Seasonality for Dr y/Artificial and Outdoor Flowers.................115 5.12 Buyer Frequency Regional Change s for Cut-Flowers and Plants..........................116 5.13 Buyer Frequency Regional Changes for Dry/Artificial and Outdoor Flowers......117 6.1 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Age.............................................123 6.2 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Gender........................................124 6.3 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Purpose.......................................126 6.4 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. ......................................127 6.5 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Income........................................128 6.6 Cut-Flowers Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Income........................................129 6.7 Ranges And Percentages of Variable Ch anges Affecting Transactions Due to Frequency of Buying for Cut-Flowers. .................................................................130 6.8 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Age..................................................................131 6.9 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Gender.............................................................132 6.10 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Purpose............................................................134 6.11 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income. ...........................................................135 6.12 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income.............................................................136

PAGE 18

xviii 6.13 Plants Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income.............................................................137 6.14 Ranges And Percentages of Variable Changes Affecting the Number of Transactions Due to Frequency of Buying for Plants. ..........................................138 6.15 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Age.............................................139 6.16 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Gender........................................140 6.17 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Purpose.......................................142 6.18 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. ......................................143 6.19 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Seasonality.................................144 6.20 Dry/Artificial Market Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Regions......................................145 6.21 Ranges And Percentages of Variable Changes Affecting the Number of Transactions Due to Frequency of Buying for Dry. ..............................................146 6.22 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Age..................................................................147 6.23 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Gender.............................................................148 6.24 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Purpose............................................................150 6.25 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income. ...........................................................151 6.26 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income.............................................................152 6.27 Outdoor Market Penetration, Buyer Fr equency and Number of Transactions Deviations From Their Means for Income.............................................................153 6.28 Ranges And Percentages of Variable Changes Affecting the Number of Transactions Due to Freque ncy of Buying for Outdoor.........................................154 6.29 Percentage of Transactions Due to Fr equency of Buying For All Flower Types..155

PAGE 19

xix Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy MEASURING DEMAND FACTORS INFLUENCING MARKET PENETRATION AND BUYING FREQUENCY FOR FLOWERS By Marco Antonio Palma Garcia December, 2005 Chair: Ronald W. Ward Major Department: Food and Resource Economics The floriculture industry is one of the fast est growing sectors of agriculture in the United States. In 1996, floriculture ranked seventh among commodity groups, behind only cattle and calves, dairy products, cor n, hogs, and soybeans. Floriculture crops, defined as cut flowers, cut cultivated green s, potted flowering plants, potted foliage, bedding and garden plants, accounted for about one third of grower cash receipts for floriculture and environmental horticulture. In or der to continue this growing trend it is of vital importance to gain insight into co nsumer preferences on floral products. Specifically, there are two key factors that n eed to be analyzed in order to understand how consumers base their decision to buy or not to buy floral products: market penetration and buyer frequency. Understandin g what are the factors that influence nonbuyers of floral products to become buyers, and the factors that influence buyers to increase their expenditures on fl oral products is vital information that the industry can use

PAGE 20

xx to design specific programs targeting differe nt demographic groups according to their specific preferences on flowers.

PAGE 21

1 CHAPTER 1 INTRODUCTION Consumption behavior has always been of great importance and a topic of focus for researchers. This importance may be attributed to the relatively strong theoretical basis for the various consumption hypotheses and an interest in empirical tests of the underlying propositions. Analysis of consumer demand has always played an important role in economic theory; this fact is evidenced by the extensive literature that exists on demand and utility (Johnson et al., 1984). The consumption of goods takes place because of the satisfaction that the goods or services provide. The consumption of tradit ional agricultural food products depends on the characteristics of the product or attribut es that can be measured or quantified. For example, milk can be measured in the quantity of calories or fat percentage. In contrast to food products, many nonfood products are consum ed because of their aesthetic value. Flowers are purchased for various reasons such as expression of love or friendship, a way to express thankfulness or a ppreciation, beautificat ion purposes for se lf, or gifts. The attributes of flowers, or more generally nonfood products, cannot be quantified; therefore the satisfaction gained from the consumpti on of these goods is closely related to the objective of the purchase. This situation also implies that the demand for these products can be influenced by the characteristics or preferences of the buyers and the reason for buying the products. This fact can be viewed during special calendar occasions (i.e., MotherÂ’s Day, ValentineÂ’s, etc) where the cons umption of floral produc ts is substantially higher compared to non-calendar occasions.

PAGE 22

2 Demand for all products depends on the charac teristics or attributes of the products. For most food products the prevai ling characteristic is to sati sfy nutritional needs. Even though flowers are not essential for survival; they possess other char acteristics that are important to food products and which influe nce the buying decision. Because flowers are not essential for survival there is a porti on of the population composed of non-buyers or infrequent buyers. Therefore there is a consid erable gap between the decision of buying or not, and this decision is based upon the demographics of the population and the occasions or periods. Understanding how cons umers make choices whether to buy or not and the perceptions of the characteristics of the products are essential to understanding the flower demand (Girapunthong, 2002). There are three groups of factors that affect the demand for floral products: external, controlled, and seasonal factors. Ex ternal factors of dema nd include inflation, wages, prices, unemployment rate, demographi c factors and other economic variables. Controlled factors of demand may be used to change perceptions and awareness by means of promotions, product developments and innovations. Even though demographic characteristics cannot be change d; the perceptions and behavi or of different demographic groups can be influenced. For example, a pr omotion program would not change the age of the consumers, but instead it can target di fferent attributes of th e products to influence purchase decisions by different age groups. Seasonal factors also affect the demand for flowers. There are certain calendar occasi ons where the demand for flowers is higher compared to other non-calendar occasions. The most common calendar occasion dates are MotherÂ’s Day and ValentineÂ’s (Ward, 1997).

PAGE 23

3 In order to analyze the demand for flowers, two types of analysis will be made. First, market penetration will be consider ed and second we will analyze buyer frequency. Because flowers are non-essential for survival in a typical month the percentage of the population that buys flowers is less than five pe rcent. From this fact arises the need to understand how consumers make the choice to purchase or not and what the factors are that influence their purchasing decisions. Afte r determining the factors that affect their purchasing behavior we can simulate and de sign specific programs to increase the entry of new consumers. Once a person becomes a consumer of flowers, and then the remaining question is the frequency of buyi ng. In an attempt to increase the total purchases it is also of great importance to understand the factor s that influence the purchasing decisions among consumers of flow ers. These two factors will be addressed in detail in a following chapter including the factors that may affect consumer responses. Overview of the Industry Floriculture has been one of the fastest growing sectors of U.S. agriculture. This sector has had a traditional average annual growth rate of about 5 percent from 1993 to 2003. However, for the first time in two decades grower sales have remained relatively flat from 2001 to 2002, with an increase of only 1.6 percent. Total floriculture sales at wholesale for large growers, that is, grower s with sales of one hundred thousand dollars or more per year, increased to almost 4.9 billion dollars in 2002, up from 3.2 billion dollars in 1996, which represents an increase of about 54 percent. From this total, there was an increase of 2.98 percent for fresh cu t flowers, 25.25 percent for potted flowering plants, 22.36 percent for foliage plants, and 73.85 percent for bedding and garden plants as we can see in Figure 1.1 (United States Department of Agriculture [USDA], 2003).

PAGE 24

4 As shown in Figure 1.2, the number of larg e growers increased from 4,566 in 1992 to a peak of 5,200 in 1998 and then decrease d to 4,741 in 2003 for a total increase of 3.83 percent. Average sales per large grower from 1992 to 2003 increased 57.66 percent for cut flowers, 47.44 percent for potted flowering plants, 91.59 percent for foliage plants, and 70.38 percent for bedding and garden plan ts. Figure 1.3 shows average sales per large grower by crop group. The total number of growers, including large growers and small growers, in 1998 was 14,308 with total greenhouse production area of 654 million square feet. For the 1993 to 1998 period, there was a decreasing trend for the total number of growers; however, the number of large growers was increasing combined with an increasing trend of production. From this f act we can see that dur ing this period, there was a transition, which the production of flor al products shifted from small producers to large growers with higher production. After 1998, the number of large growers also has had a decreasing trend in combination with an increasing production trend. This decline in the number of growers has been attributed to increased import competition and consolidations to achieve ec onomies of scale, such as co ntract production with large retail chains (Schum acher et al., 2000). The value of total U.S. imports of floric ulture and nursery products increased from 712.4 million dollars to 1.2 billion from 1994 to 2003 as shown in Figure 1.4. The countries from which the U.S. imports the most are Colombia, Canada and the Netherlands for a combined value of 908.8 m illion or 72.72 percent of total imports in 2003 (Figure 1.5). Cut flowers represented 59 and 49 percent of the imports for 1994 and 2003, respectively. The total value of cut flow er imports increased 45 percent from 420 million in 1994 to 611 million in 2003.

PAGE 25

5 Figure 1.1 Large Grower Sales by Crop Group From 1985 to 2003. Source. Economic Research Service, USDA, 2003. Figure 1.2 Total Number of Large Growers in the US From 1992 to 2003. Source. Economic Research Service, USDA, 2003. 1985198619871988198919901991199219931994199519961997199819992000200120022003 0.20 0.40 0.60 0.80 1.00 Large Grower Sales. Millions Cut FlowersGreen House Dry/Artificial 199219931994199519961997199819992000200120022003 4200.00 4400.00 4600.00 4800.00 5000.00 5200.00 5400.00 Number of Large Growers.

PAGE 26

6 Figure 1.3 Average Sales Per Large Grow er by Crop Group from 1992 to 2003. Source. Economic Research Service, USDA, 2003. Figure 1.4 US Total Imports and Exports from 1976 to 2003. Source. Economic Research Service, USDA, 2003. 199219931994199519961997199819992000200120022003 0.00 0.20 0.40 0.60 0.80 Average Sales per Large Grower. Millions Cut FlowersGreen House Dry/ArtificialOutdoor 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 Million $ ImportsExports

PAGE 27

7 Figure 1.5 US Imports of Cut Flowers and Nursery Crops by Country, 2003. Source. Economic Research Service, USDA, 2003. Figure 1.6 US Exports of Cut Flowers and Nursery Crops by Country, 2003. Source. Economic Research Service, USDA, 2003. Canada 26.4% Colombia 27.8% Netherlands 18.5% Others 27.3% Canada 52.9% Netherlands 19.8% Other 27.3%

PAGE 28

8 The major exporter of cut flowers to the U.S. is Colombia with 343.6 million in 2003, for 56.21 percent of the total U.S. cut fl owers imports. In terms of nursery stock, the major exporters to the U.S. are Canada a nd the Netherlands with a combined value of 473.6 million or three quarters of the total U.S. nursery stock imports. Total U.S. exports of cut flowers and nursery stock increased from 252 million to 272 million from 1994 to 2003. About 53 and 20 percent of U.S. floric ultural exports are to Canada and the Netherlands, respectively, as shown on Figure 1.6 (USDA, 2003). Even though fresh cut flowers, potted flower ing plants and dry-ar tificial flowers are fundamentally different and substitutable to so me degree, there are certain similarities of attributes among these products if we analyze them in terms of the purpose of use. They can be used to express love, thanks, re flect emotions, project beauty, and show environmental concerns. Consumer expenditure patterns may change among these products even though they are physically differe nt. These consumer patterns are affected by many factors, including income, purpose of use, occasions, information, perceptions, and sources for purchases. The level of c onsumer expenditures de pends on three basic components: market penetration, frequency of transactions among buyers and prices (Girapunthong, 2002). Demand analyses fo r floral products differ among other agricultural commodities in the sense that fo r other agricultural commodities the quantity consumed is used directly in the analysis. In the specific case of fl owers, a consumer may purchase a single stem rose or an arrangem ent. Therefore demand studies on flowers generally replace quantity observed by the numbe r of transactions given in a defined period of time.

PAGE 29

9 Given the reasons why consumers buy flow ers and the fact that flowers are nonessential for survival, one can observe that the consumption of floral products has significant fluctuations over time. These fl uctuations result in a higher number of transactions during special calendar occasi ons during the year. The most important calendar occasions are Valen tineÂ’s Day, MotherÂ’s Day, East er/Passover, Thanksgiving, and Christmas/Hanukah. The American Flower Endowment consumer tracking study of 1999-2000 reported that Christmas/ Hanukah had the highest numb er of transactions, and represented one-third of all calendarÂ’s oc casions; MotherÂ’s Day, Easter/Passover, ValentineÂ’s Day and Thanksgivi ng accounted for 20, 18, 16 and 5 percent respectively. It is of vital importance that any demand analysis for flowers takes into consideration these special calendar occasions. In the last decade, the industry has e xperienced many changes, including industry programs adopted to increase the total demand for flowers. Brand a nd generic programs have been adopted to entice the demand for floral products. The fundamental difference between brand and generic advertisement pr ograms is that brand advertisement is adopted by a firm to benefit sp ecifically that firm while generic advertising seeks to increase the demand of the whole flower indus try. Examples of generic advertisement in the flower industry include PromoFlor and the Flower Promotion Organization (FPO) promotion programs. PromoFlor was a generic promotion program implemented in 1993, which had the main objective to expand the total demand for fresh cut flowers and greens. Ward (1997) showed that the PromoF lor program was successful in increasing the expenditures per buyer, generating a net gain of 5.6 dollars of additional revenue for every one dollar spent on the promotion prog ram at the wholesale level. However,

PAGE 30

10 PromoFlor was terminated in June 1997. Rimal (1998) explained that one possible reason for the termination of the program was equ ity concerns regardi ng the distribution of benefits among the fresh cut flower system. The FPO implemented a promotion program that targeted five U.S. citi es with the goal of increasing the frequency of buying fresh cutflowers by existing female flower buyers for non-calendar occasions. Potential changes in market penetration and buying frequency were adopted as measurement criteria for judging consumer responses to the promotions. The program had a positive impact on attracting new buyers and increasing buyer expe nditures per transaction. For every dollar spent on the promotion program there wa s an additional $9.5 dollars generated on expenditures of fresh cut flowers in the target area (Ward, 2004). In addition to promotion programs, the de velopment of new technologies, such as the Internet, has made possible the creation of new sources for buying floral products. Examples of Internet-based firms include FTD, 1-800-Flowers, and Teleflorist. These businesses have created product diversificat ion such as floral baskets and bouquets, which can influence the purchasing decisi ons of consumers and their tastes and preferences. The main objective of these fi rms is to increase th e demand for floral products through the use of technology and make the services more convenient (Girapunthong, 2002). Problem Statement The floriculture industry is one of the fast est growing sectors of agriculture in the U.S. In 1996, floriculture ranked sevent h among commodity groups, behind only cattle and calves, dairy products, cor n, hogs and soybeans. Floricu lture crops, defined as cut flowers, cut cultivated greens, potted flowering plants, potte d foliage, bedding and garden plants, accounted for about one third of gr ower cash receipts for floriculture and

PAGE 31

11 environmental horticulture (Schumacher et al ., 2000). In order to continue this growing trend it is of vital importance that one obtains insight into consumer preferences on floral products. Specifically, there are two key factors that need to be analyzed in order to understand how consumers base their decision to buy or not to buy floral products: market penetration and buyer frequency. U nderstanding what th e factors are that influence non-buyers of floral products to b ecome buyers, and the factors that influence buyers to increase their expenditures on floral products, is vital information that the industry can use to design specific program s targeting different demographic groups according to their specific preferences on flowers. Objectives The general objective of this study is to analyze the factors that drive the demand for flowers in the U.S. in terms of market penetration and frequency of buyers for cut flowers, potted flowering plants, dry/artif icial and outdoor plants. Three specific subobjectives will be accomplished in order to achieve the overall main objective: 1. Given the number of buyers and househol ds, analyze the factors that attract nonbuyers of floral products to become a buyer, which is market penetration. 2. Examine the factors that contribute to in creasing consumer expenditures on flowers depending on the type of product, source, reason for buying, seasonal considerations and demo graphic characteristics. 3. Use the results from the market penetra tion and buyer frequency models to make simulations on specific combinations of the product attributes and demographic characteristics, to rank the importance of those factors impacting both market penetration and frequency. Research Methodology Our data set shows the number of buyers and the number of households, which would, allows one to calculate the market pe netration ratio. Also the data have showed

PAGE 32

12 total expenditures, number of buye r transactions and quantities. As stated before, in the case of flowers, it is no t of much advantage to use the qu antity of purchases because that it is hard to record whether a quantity of one means one single stem rose, or an arrangement of multiple flowers. That is w hy the number of transactions is replaced by the quantities. The frequency of buying can be calculated by dividing the number of transactions by the number of buyers. Also pri ce can be calculated by simply dividing the expenditures by the number of transactions. Th e focus of the study is divided into two main parts: market penetration m odels and buyer frequency models. Based on these main variables mentioned in the first paragraph, market penetration models are used to analyze what factors in fluence consumers to become a buyer or not. This model is one of the main two topics of this study; the second pa rt will address buyer frequency models. When a consumer has beco me a buyer, that is, households with at least one transaction, what are the factors that influe nce how much consumers buy. Because both models, market penetration and buyer frequency, have a cluster of observations on the lower limit, a model is needed that will take into account its asymptotic distribution. The market penetra tion model has a lower limit at zero, while the buyer frequency has a lower limit of one, since in order to be defined as a buyer a household must have made by definition at le ast one transaction per month or more. The model that deals with this type of clustering of the data is the Tobit model. A Tobit model combined with frequency of buying models w ill be used in order to analyze the factors that affect the number of transactions in a given period of time. After the implementation of market pene tration and buyer frequency models, the results can be used to simu late different products, reasons for buying, outlets sources and

PAGE 33

13 demographic characteristics such as gender, age, income groups, etc. These simulations will result in specific recommendations for the flower industry in an attempt to increase the overall demand for flowers in the U.S. Data and Scope Data used in this study are aggregate data collected by the National Panel Diary Group (NPD) and sold through the American Flower Endowment. The data provide statistics on behavior of consumers includ ing transactions and expenditures on flowers for both gift and self-use. NPD data were available from consum ers purchasing diary completed by households from a large demogra phically representative sample of U.S. households. The data have two separate va riables for number of households and the number of households that buy flowers. This separation between household buyers and non-buyers would allow the calculation of the ma rket penetration ratio, or the percentage of the overall sample that ar e buyers of floral products. Mo nthly purchasing data from July 1993 to June 2004 will provide information regarding the number of households buying flowers, the total number of households, expenditures, number of tran sactions and quantity for both gift and self-use of flowers. Flower data used in this analysis are cat egorized into four different income groups: under $25,000; $25,000-$49,999; $50,000-$74,999; and $75,000 or more. These income groups have data from five ma in categories: product form, purpose of the purchase, time, demographic characteristics, and geographic lo cation. Product form refers basically to the four sub-categories: cut flow ers, potted flowering plants, dry/artificial a nd outdoor; also as subcategories of cut flow ers there are arrangements, single stem/ bunches and others. The purpose of the purchase is either for self -use or for gifts. The time of the purchase (calendar vs. non-calendar occa sions) is recorded along with demographi c characteristics

PAGE 34

14 including income, education, employment, occ upation, family size, gender, marital status and regional location. Organization of the Study This study has seven chapters. Chapter 2 is a review of literature on demand theory, floral demand analysis and literatur e on market penetrati on and buyer frequency. Chapter 3 consists of descrip tive statistics of consumer demand for the flower industry in the U.S. with expenditures, transactions number of households, reasons for buying, product form and demographic characteristic s being addressed. Chapter 4 presents a conceptual framework for market penetratio n and frequency theory and measurement. Chapter 4 is basically where econometric models and the development of estimation techniques will be constructed. Chapter 5 includes the model specification and estimation. In this chapter th e results are discussed and interpreted. Chapter 6 is simulation analysis and has two main sec tions: (1) ranking of market penetration variables and ranking of frequency of buying; and (2) projections: using demographic trends. Finally, Chapter 7 gi ves a summary of the study.

PAGE 35

15 CHAPTER 2 LITERATURE REVIEW This chapter consists of four sections. First, the consumer behavior will be addressed with an emphasis on preferences and choices. The properties of preferences will be listed and described. Indifference curves marginal rate of substitution and utility maximization will be discussed. Second, consumer demand analysis will be addressed and discussed. The consumer allocation problem and demand properties will be the main focus of this section. Third, different marke ting models will be pr esented and discussed. These models are the fundamental theory be hind market penetration models and buyer frequency models. The two main parts of this section are: market penetration models, where different functions for market penetra tion models will be de scribed verbally and mathematically; and frequency of buying models, where we will introduce the notion of repeat-buying and will describe the model us ed to analyze buyer frequency models. And finally, past studies on th e flower industry will be listed and summarized. Consumer Behavior Consumer behavior is often presented in two ways: in terms of preferences or in terms of possibilities. In discussing consum er behavior, generally the main focuses are preferences, axioms of choice and utility functions and their properties. Unlike preferences, choice as an opportunity is often directly observable so that, to the extent that variations in behavior can be traced with variations in opportunities, one has a straightforward and objective expl anation of observed phenomena.

PAGE 36

16 Preference is a key factor in consumer or buyer behavior. Preference determines choices made by buyers of most products and serv ices within the limits of a set of defined constraints. Buyer preference will determin e what goods or attributes of a good are selected and will also determine the selection of one good over another. An understanding of this concept is vital to th e market development of a product such as flowers, a product consumed for its aesthetic value. Considering all factors, when a consumer re ports that “A is preferred to B”, then that particular consumer feels better under situation A than under situation B. This preference relation is assumed to have three basic properties: completeness, transitivity and continuity. The completeness property states that if A and B are any two situations, the individual can always specify one and onl y one of the following possibilities: (1) A is preferred to B, (2) B is prefe rred to A, or (3) A and B are equally attractive. Individuals completely understand and can always choose th e desirability of a ny two alternatives A and B. The transitivity property infers that if a consumer reports, that “A is preferred to B” and that “B is preferred to C,” then the c onsumer must also report, that “A is preferred to C.” Therefore, the transiti vity property assumption states that individual choices are internally consistent. The conti nuity property indicates that if an individual reports, that “A is preferred to B,” then under simila r accommodated circumstances, the consumer must also report that “A is preferred to B” (Nicholson, 1998). Given the assumptions of completeness, tran sitivity and continuity, it is possible to establish that the consumers are able to rank order all possible choices from the least desirable to the most desirable. The ordering implies an underlying level of utility or satisfaction that is derived from each choice. If an individual has a preference of choice A

PAGE 37

17 over choice B, then the utility obtained from A, denoted by U(A), is greater than the utility derived from B, U(B). Utility can be expressed in ordinal number values, where the higher the value the higher the utility. These values will reflect the preference ordering for a set of choices. Because utility refers to overall satisfac tion from the consumption of a product or a service, it is influenced by a variety of factors. A consumerÂ’s utility is influenced by the following factors: consumption of physical commodities, income, peer group pressures, personal experiences, and the general cultu ral environment (Nicholson, 1998). Individual preferences or utility are assumed to be represented by a function of the form (2.1) ) ; ,..., , (3 2 1other X X X X U utilityn where N iX, 1 refers to factors or produc t attributes and the other are variables that affect the utility for the product such as demographic characteristics. A consumer preference mapping can be shown graphically through the use of indifference curves. An indifference curv e represents all combination of product attributes that provide the same level of satisfaction (utility). For simplicity, one can assume that there are only two product attributes for a produc t, attribute A and attribute B. Then the utility for the product would be re presented as U(A,B). Thus an individual is indifferent among the combinations represente d by the points gra phed on the indifference curve (Pindyck and Rubinfeld, 2001). Figur e 2.1 is a graphical representation of consumer preference mapped by an indifference curve. In the figure, there are only two products A and B. For a particular consumer, the same level of utility is attained at both points A and B (note those are points along the same indifference curve and therefore provide the same level of u tility. Hypothetically a

PAGE 38

18 consumer would be willing to give up some u tility obtained from product A in order to increase the utility received from product B, and vice vers a. In other words, it is suggested that a consumer would be willing to receive a product with a less favorable amount of B in order to obtai n more quantity of product A, or a consumer would be willing to accept less of product A in or der to obtain more of product B. Figure 2.1 Graphic Presentation of Indifference Curves This situation is true if all other things remain constant (ceteris paribus assumption). An individual receives the same total utility by consuming the combination A*,B* (point A) or the combination 2 2, B A, (point B). Again, the curve represents all the consumption bundles that provide the same level of utility or stated in a different manner, all the bundles that the individual ranks equally in terms of derived utility. However the consumption would take place at point A becau se given the budget line that would be the

PAGE 39

19 point where utility would be maximized a nd the budget constraint would hold. The slope of the indifference curve is negative, meaning that if an individual is willing to give up an amount of product A, then he must be compen sated with an increase in the amount of B in order to remain indifferent between th e two bundles. This compensation represents a trade-off between products A and B. This negative slope of the indifference curve is called the marginal rate of substitution (MRS) at that point, and it is expressed as follows: (2.2) 1I U dB dA MRS where the notation means that the slope has to be calculated along 1I indifference curve. The slope or the MRS represents the trades a consumer would voluntary make. Utility increases as the indifference curve shifts out due to other constraints being relaxed. This fact can be seen in Figure 2.1, as ) (1I U < ) (2I U < ) (3I U In order to maximize the overall utility, a consumer would be located on the indifference curve situated as far from the origin as possible within the context of the constraints the consumer faces. In Figure 2.1 the budget line establishes th e combination of products along the indifference curve. The budget line is the constraint that a consumer faces as a result of relative prices or costs and a fixed level of income. In other words, the budget line, which is simply the ratio of the prices of product A and B, repres ents all combinations of A and B for which the total amount of money spent is equal to consumer income. Consumers maximize utility by choosing a bundle that is on the indi fference curve and that is tangent to the budget line.

PAGE 40

20 Consumer Demand Analysis There exists extensive literature in de mand analysis. Some of the early works documented on demand analysis include Hi cks (1946) and Samuelson (1947). Other demand analysis studies that are often us ed for pedagogic purposes include Goldberg (1967), Phlips (1974), Powell (1974), Theil a nd Gabrielson (1975, 1976) Barten (1977), and Deaton and Muellbauer (1980). The formul ation of the utility maximization problem implies the existence of a utility function w ith specific properties. The purpose of demand analysis is to study from the utility func tion to the demand function and analyze the properties these functions ha ve (Johnson et al., 1984). Utility Maximization and Demand Functions The previous section explained the assu mptions about consumer behavior. These assumptions are introduced into consumer demand analysis through the specification of a utility function. The utility function measures the level of satisfact ion a consumer would obtain from a product or service on a given period of time. The utility function is denoted by the function: (2.3) ) (iq u u where the expression means that the utility is a function of the levels of consumption at period i. The utility is represen ted by an ordinal number with no definite scale; the higher the number the higher the level of utility ob tained from the consumption of a bundle of goods or services. The utility function defined in 2.3 has three basic assumptions. First, it is assumed to be strictly increasing. A function f(x) is said to be strictly increasing on the interval (a,b) if

PAGE 41

21 (2.4) ) ( ) (1 2x f x f whenever 1 2x x Second, it is assumed that the utility function is quasi-concave. A function f(x) is quasiconcave over the in terval (a,b) if (2.5) )] ( ), ( min[ ] ) 1 ( [2 1 2 1x f x f x x for all 2 1, x x in (a,b) and all 1 0 The function f(x) is strictly quasi-concave if th e strict inequality (>) holds. Third, and last a utility function is assumed to be twice co ntinuously differentiable. A function f(x) is continuous at the point x=a if the following conditions hold: (2.6) ) ( x fa x and f(a) exist and ) ( ) ( a f x fa x The interpretations of these conditions are easie r if the indifference curve is used (Figure 2.1). Strict quasi-concavity of the utility f unction (2.3) ensures that the indifference curve does not contain linear segments or bends back on themselves. The assumption that the utility function is strictly increasing implie s that a consumer prefers more to less. The twice continuously differentiable assumption as sures that the indifference curves are well defined and not kinked. Marginal utilities are the first partial derivatives of the utility function. These marginal utilities should be interpreted as the increase in total utility that takes place because of the consumption of an addition al unit of the commodity per unit of time. These derivatives of the utility function are represented by: (2.7) 0 i iq u u (i=1,2,3,Â…,n), and they are positive because of the continuity and the differentiability assumptions; the second partial derivatives of the utility func tion are, by YoungÂ’s theorem, symmetric, and they are represented as

PAGE 42

22 (2.8) ji i j j i iju q q u q q u u 2 2 (i,j = 1,2,3,Â…,n). The second partial derivatives of the utility function can be interpre ted as the rate of change of the first partial derivatives and ther efore the behavior of the marginal utility for each of the commodities. In order for the analys is to be complete it must include how the marginal utilities change with changes in the consumption levels of other commodities, that is, iju, when j i. The utility function (equation 2.3) is maximized subject to a budget constraint, that specifies that consumers have no savings, or in other wo rds, that consumers spend all of their income: (2.9) m q p ', where ) (ip p is the n-element column vector of the prices and m is consumer income. Hence, our maximization problem is: (2.10) Max ) (iq u u Subject to: m q p '. The maximization of the utility function subject to a budget constraint in equation (2.10) is carried out by the Lagra ngian method. According to this method the Lagrangian expression is formed as follows: (2.11) ) ( ) ( ) ( m q p q u q L where is the Lagrangian multiplier and it is interpreted as the marginal utility of income. Differentiating the Lagrangian equa tion (2.11) with respect to each of the arguments, iq and yields the first or der conditions (FOC): (2.12) 0 p uq and 0 m q p

PAGE 43

23 where qu is the vector of derivatives of the util ity function with respect to the quantities, ) ,..., 3 2 1 ( n i qi The second order conditions of the Lagrangi an for a maximum can be expressed in the following manner: (2.13) 0 Ux x for all x such that 0 x p The matrix U is the Hessian matrix, and co ndition (2.13) is assured because of the assumption that the utility function U is strictly quasi-concave. The system of equations can be solved for nq q ,...,1 and in terms of prices and income. The resulting unique expressions are: (2.14a) ) ,..., (1m p p q qn i i (2.14b) ) ,..., (1m p pn The demand function iq is of extreme importance in both theory and practice, because it describes how a consumer will behave when f aced with an alternativ e set of prices and a particular income. The term shows that the marginal utility of income depends on both prices and income. Economists often find it more convenient a nd useful to express the restrictions on the demand system in terms of elasticities ra ther than derivatives (Johnson, et al, 1984). Elasticity of demand can be calculated from the maximization procedure described above. There are three basic el asticities of demand: (1) ow n-price elasticity, (2) crossprice elasticity, and (3) income elasticity. The own-price elasticity of demand can be defined as the proportionate change in the qua ntity purchased to the proportionate change in its own price as follows:

PAGE 44

24 (2.15) 0 i i i i iiq p p q. If 1 ii, then there is a unitary elasticity of demand, which means that a 10 percent increase in price will decrease the quantity consumed on the same proportion, in this case 10 percent. If the elasticity of demand is gr eater than one in magnitude (absolute value), then the demand is price elastic, because th e percentage decline in quantity demanded is greater than the percentage increase in price. If 1 0 ii, then, the demand is inelastic. The cross-price elasticity of demand of a good is the responsiveness of the quantity of that good to change s in prices of other goods: (2.16) i j j i ijq p p q If the cross-price elasticity of demand is negative (0 ij ), then good i and good j are complements, while if the cross-pri ce elasticity of demand is positive (0 ij ) then the two goods are substitutes. The income elasticity of demand can be defined as the proportionate change in quantity purchased relative to changes in in come, given that prices are held constant: (2.17) i i imq m m q If the consumption of a good incr ease with income in creases, that is the income elasticity of demand is positive, then these types of goods are defined as normal goods. If the marginal propensity to spend on iq is less than the average propensity to spend on iq (1 0 im ), then these types of goods are refe rred as necessity goods. If the marginal

PAGE 45

25 propensity to spend on iq is greater that the average propensity to spend on iq (1 im ), then the good is known as a luxury good. Up to this point the maintained hypothe sis of demand theory is that a consumer will select from the set of affordable comm odity bundles the one that would yield that consumer the maximum possible utility attain ed. The assumptions and results of this hypothesis have been explained above in the utility maximization framework. A different approach to analyzing the consumer optimi zation problem can be used by applying the duality concept. Duality was introduced by Hotteling (1932) and developed and popularized by Shepard (1953) with his wo rk on production and cost functions and employed to advantage in the analysis of consumer demand. The duality formulation of the consumer allocation problem is developed from the expression: (2.18) m q p Sto q u m p vq : ) ( max ) ( where v(p,m) is the indirect utility function. The indirect utility function is defined as the maximum attainable utility level for a given set of prices and a particular income. The indirect utility function can be used to obtain the direct utility func tion. If the indirect utility function v(p,m) is minimized with resp ect to prices and income and subject to the budget constraint, the direct util ity function u(q) is obtained. This property of the direct and indirect utility function is of great importance for analy tical purposes. In order for this relationship to exists, the indirect util ity function v(p,m) has to have the following properties: (1) continuous, (2) decreasing in prices, (3) increasing in income, (4) strictly quasi-convex in prices, and (5) homogeneous of degree zero in prices and income.

PAGE 46

26 The advantage of using this approach in analyzing the consumer demand problem is that it is easier to derive the demand f unctions and associated conditions. For instance, the uncompensated (Marshallian) demand func tion for the ith commodity can be obtained by differentiating the indirect utility function in (2.18) with respect to prices and income and applying RoyÂ’s identity: (2.19) *) ( ] ) ( [ ] ) ( [i i iq m p q m m p v p m p v (i=1,2,3,Â…,n). In a similar way the uncompensated demand function can be obtained by applying the Hotteling-Wold identity to the direct utility function (Johnson, et al, 1984): (2.20) *) ( / ) / (i i j i j ip m q p q u q m q u (i=1,2,3,Â…,n). Properties of the Demand Functions Restrictions on consumer demand functions or properties are obtained from the manipulation of the first order conditions (FOC ) specified in (2.12). Properties of demand are developed by considering the consequences of shifts in th e first order conditions with respect to prices and income. The results of these shifts are described by the partial derivatives of the first order conditions for both prices and income. There are four basic demand properties or restricti ons: (1) adding-up restriction, (2) symmetry restriction, (3) homogeneity restriction and (4) negativity restriction. Adding-up restriction The adding-up restriction of the de mand function is derived from the monotonicity assumption of preferences and the budget constraint. The adding-up restrictions has two main components know n as the Engel and Cournot aggregation restrictions. The Engel aggreg ation restriction relates to income elasticities and it is

PAGE 47

27 represented as the sum of the income elas ticities weighted by its expenditure share, m q p wi i i/ (2.21) i im iw 1 The Cournot aggregation restriction states th at the sum of the cross-price elasticities weighted by its expenditure shares has to e qual the negative of the expenditure share of good j: (2.22) j i ij iw w where j=1,2,3,Â…,n. Symmetry restriction The symmetry restriction states that th e matrix of compensated cross-price substitution effects is symmetric. This is true because of the continuity and differentiability assumptions of the utility function that follow YoungÂ’s theorem. (2.23) jm i j ji j im j i ij iw w w w w w or ji ij for j i where ij is the cross-price elasticity, iw is the weighted expenditure share of good i, im is the income elasticity for good i, and im j i ij i ijw w w is the Slutsky coefficient for good i and j. Negativity restriction The matrix conformed by the second deriva tives of the expenditure function, by ShephardÂ’s lemma, constitutes a substitution ma trix and hence should be a negative semidefinite. For the necessary condition, the diagon al elements of this matrix would have to be negative, which can be reformulated in elasticity form by using the Slutsky equation and setting i=j as well as multiplying by i iq p/: (2.24) 0 im i iiw

PAGE 48

28 Homogeneity restriction The homogeneity restriction of the demand f unction states that the sum of all the direct and cross-price elasticities for a partic ular commodity i, is equal to its ith income elasticity: (2.25) j im ij where i=1,2,3,Â…,n. Marketing Research Models Marketing is the process by which syst ems of consumption are developed and changed by individuals and orga nizations. A marketing system consists of organizations that provide goods and services to consumer s, the flows of information between the organization and the consumer, and the physical flow of goods and services (Parsons and Schultz, 1976). The main focus on marketing res earch aims to answer questions in regard to the effects of market ing instruments and specif ic household demographic characteristics on various marketing performance measures. Some examples of these marketing measures include sales, market shares, brand choice and inter-purchase times. With the results on these measures it is possi ble to select which marketing instruments will be used targeting specific household group s. In recent years there have been major advances in the data collection methods allowing the resear chers to access large databases on individual consumer habits. In addition to this through the use of supermarket loyalty cards and scanners researchers can trac k what individuals purchase; hence the researcher gains information not only on stated preference, that is what the consumer reports, but also revealed preferen ce, that is what the consumer actually purchased. The large amount of marketing data implies that simple graphical tools and elementary modeling techniques in most cases are not sufficient to analyze and explain

PAGE 49

29 present day marketing problems (Franses and Paap, 2001). In order to obtain results that are accurate with reality researchers need to use the available information to build marketing models that explain and adapt to re al life. This research project will use two marketing models: market penetratio n models and buyer frequency models. Market Penetration Models The market penetration process refers to individuals that are non-buyers of a product or service and then when given some type of information-based decision they become buyers. The main question on market penetration is what are the factors that attract new buyers for the product. In the case of flowers, there has been a change in the trend of flowers. Advertising campaigns ha ve shown that flowers are not only a product for gifts, but also for self use for decora tion purposes and personal enjoyment. One can use the available data that includes reasons fo r buying, transactions, prices, time of the year (calendar vs. non-calendar occasions), and demographics, to segment the population and identify the factors that mo st likely attract non-buyers of floral products to become buyers. There are many mathematical functions used on market penetration models; this section will present and discuss the most important mathematical functions used in market penetration. The main source for this section was th e market penetration functions developed by Fleck (1981). Logistic function The logistic function can be expressed in the following differential equation: (2.26) )] ( ~ )[ ( ) ( t y y t y t y where t t y t y ) ( ) ( is the increase in demand at period t, y(t) is the demand accumulated at period t,

PAGE 50

30 is the positive factor of proportionality, y ~ limit of saturation ) 0 ~ ( y The interpretation of this differential equa tion is quite simple. The increase in demand yÂ’(t) is determined by the market potential )] ( ~ [ t y y which is not exploited yet at time period t and by the stock y(t) reached in t. At the beginning of the penetration process the demand increase is small, because even though there is a high potential market, few consumers will accept the product. With the in creasing number of buyers, the potential market decreases and ) ( t y also increases. Then when th ere is a large enough portion of new buyers of the product, ) ( t y decreases because the po tential market becomes smaller. With a positive demand at period t, the so lution to the differential equation (2.26) can be obtained by the separation of the va riables and decompositi on of a fraction into partial fractions: (2.27) bt ae y t y 1 ~ ) ( One of the criticisms of the l ogistic function is that it de scribes the process of market penetration as an endogenous pr ocess. Because of this fact, the parameters are constant meaning that they are not affected by external market influences. The Pyatt function This model combines both exponential an d logistic growth. Exponential growth with limit saturation an d with no initial resistance to buy describes products that are frequently bought (routine purchases). Similarly, this situation also applies to purchases caused by new trends, where at the beginning of each period the targeted consumer

PAGE 51

31 groups are already well informed about the product and the desire to buy increases. The foundations of the Pyatt model are base d on the following differential equation: (2.28) )) ( ~ )( ( )) ( ~ ( ) ( t y y t y t y y t y The first term is the exponential growth and the second is the logist ic growth. Depending on the value of the shape of the curve will be different. If 0 then the logistic function is obtained. If 1 the logistic behavior is dominant. If 1 then a hyperbolic function is obtained. The Pyatt cu rve is obtained as the solution of the differential equation (2.28) with the initial condition y(0)=0 : (2.29) y e e t yt t~ 1 1 1 ) () 1 ( ) 1 ( The Gompertz function This function was first applied by the economist Gompertz in 1925 for the analysis of mortality. The Gompertz function can be expressed as follows: (2.30) tbce y t y ~ ) (, where both b and c are constant and b>0 and 0
PAGE 52

32 fact that the penetration pro cess slows down as the degree of saturation increases. The above Weblus model can be expressed in an equation of the form: (2.32) )) ( 1 )( ( ) ( t y t y t a t y Here the constant term described on the logistic model (2.26) is replaced by the time dependent parameter a/t. In this equation, Weblus uses as the limit of saturation ) ~ ( y the relative limit of 1, which corres ponds to the total market or a market share of 100 percent. The solution to the differential equation (2.32) where 2 1 ) (0 t y reads: (2.33) at t t y 01 1 ) (, where 0t is the period of time between the beginn ing of the penetration process and the point when the saturation reaches one half or 50 percent of market share is attained. The log-inverse function Prais and Houthakker applied this func tion for the first time when analyzing family budgets in 1955. This function is define d by a growth rate, which is determined in the following manner by the stock level y(t) and time squared: (2.34) 21 ) ( ) ( t t by t y where b is a positive constant, and then a solution is obtained where y a ~ ln for t : (2.35) t b ae t y ) (.

PAGE 53

33 The disadvantage of the log-i nverse function is that it only possesses one degree of freedom or in other words, one parameter, and then it can be difficult to adjust this model to real data. The lognormal function The logistic and the cumulative normal di stributions are very similar. They both possess a disadvantage in that both have a prope rty that one half of the limit of saturation is obtained at the point of infl ection. Then, in order to obtain a function that is similar but without that property the lognormal function was defined as follows: (2.36) td e y t y0 ) (log 2 12 22 1 ~ ) ( where is the standard deviation and is the mean of the distribution. Repeat Buying Models Repeat buying is when a consumer buys a product more than once in a given period of time. Consumers have pre-purchas e needs, perspectives, attitudes, the experience of previous usage, and external in fluences, such as advertising and promotion programs, retail availability, personal selling and word of mouth eff ects, and differences in products, services and prices. The consumer has to make decisions regarding what products to buy and at what prices and where to buy the pr oducts. All of these characteristics form a post-buying experien ce in the customerÂ’s mind after the purchase takes place; based on all these factors a c onsumer would choose depending on the level of satisfaction or utility obtained from the pr oduct or service whethe r to re-purchase the product or not. There are basica lly three cases of repeat buying situations that can be defined. First, if a consumer buys more than one product in one or more purchase occasions in the given time period. In this cas e, consumers differ in how often they repeat

PAGE 54

34 buying the products. The frequency of buying w ould be 0 for a consumer that did not purchase the product and 1 for consumers that purchased the product on ce. For the repeat buyers the frequency will be 2, 3, 4, et c., depending on the numbe r of repeat buying occasions they purchased the product. Th e second way of repeat buying refers to consumer that may buy the product in more than one time period. Then a model can be formulated for repeat buying behavior under stationary and no trend conditions. The third and last form of repeat buying behavior is that more than one unit may be purchased on the same purchase occasion (Ehrenberg, 1988). In the case of the data set, the data are organized in such a way that we do not have information about the non-buyers. The frequency of buying is then the number of purchases a household made in a given period of time (months). The frequency of buying of flowers is affected by external seasonal f actor. As an example, the frequency of buying as well as the total number of buyers increa se during special calenda r occasions such as MotherÂ’s Day, ValentineÂ’s, Christmas, etc. In order to analyze the frequency of buying, and because of the fact that the data set does not contai n any information about nonbuyers of floral products, we then will employ the Tobit model. The Tobit model is an ex tension of the Probit mode l. The Nobel Prize winner James Tobin developed the Tobit model in 19 58. Because the only available data is on consumers that did purchase floral products then the Tobit model will be the most appropriate model to make the analysis. Consumers are divided in two groups, one consisting of 1n consumers about whom we have info rmation on the regressors (income, education level, gender, etc.), as well as the regressand (expenditures on flowers); and the second group, 2n consists of consumers about whom we only have information about the

PAGE 55

35 regressors. A sample consisting of people on the second group is known as a censored sample. That is the reason why the Tobit model is also known as censored regression models or a limited dependent va riable model (Gujarati, 1995). The general mathematical formulation of the Tobit model can be expressed in the following equation assuming the constraint from below: (2.37) '*i i ix y 0 iy if 0*iy i iy y if 0*iy There are three different conditional functions that can be analyzed depending on the purpose of the study. For the index variab le, sometimes called latent variable, ] [*iy E is ix However, if the data are always censored this result would not be useful. Then for any observation randomly drawn from the samp le, which may be censored or not, the expected value is: (2.38) ' ] [i i i i ix x x y E where, / / 'i i ix x The estimation of the model is done th rough the use of the maximum likelihood technique. The log-likelihood function for the censored regression model is: (2.39) 0 0 2 2 2' 1 log log 2 log 2 1 logi iy i y i ix x y L The two parts correspond to the classical regression for the non-limit observations and the relevant probabilities for the limit obser vations, respectively. This log-likelihood

PAGE 56

36 function can be simplified following OlsenÂ’s reparameterization, where / and / 1 to obtain: (2.40) 00 2 2) ( 1 log ) ( log ) 2 log( 2 1 logi iyy i i ix x y L Review of Past Studies on Flower Products The floriculture industry has evolved rapi dly in recent years. The introduction of mass-market retailers such as supermarkets department stores and Internet-based business has changed the marketing paradigm of floriculture. Compared to the other food products such as milk, meat, citrus, etc., floriculture lacks an extensive marketing literature. Early studies focu sed on consumer preferences toward fresh cut flowers. The following section presents some studies reporte d on floriculture that will be relevant to the analysis of this research. Miller (1983) performed an extensive subsector analysis for the fresh cut-flower industry in the United St ates. The methodology employed by Miller was to use an approach that followed the structure, conduc t and performance framework in order to analyze the existing conditions of the industry a nd to predict future trends. In his analysis, Miller observed that there were special cal endar occasions when the demand for flowers was substantially higher, and other noncalendar occasion where the demand was decreased substantially. He al so determined that the demand for flower arrangements was inelastic, meaning that consumers are not highl y responsive to changes in price of flower arrangement products. Also Miller observed a nd reported a rapid growth of mass-markets in the fresh cut-flower industry and predicted that they would become more important in the future.

PAGE 57

37 Tilburg (1984) in his st udy titled “Consumer Choice of Cut Flowers and Pot Plants” analyzed consumer panel data of households in the Netherlands. The main objective of the study was to relate aspects of consumer behavior on cut flowers and pot plants to marketing variables and demogr aphic characteristics of households, to determine whether market segments exists or not, and to determine the feasible application of marketing models in the flow er industry. He identified three market segments: the first segment consisted of 44 pe rcent of the households and was sensitive to prices but insensitive to national advert isement; the second segment consisted of 40 percent of the households, and was insensitiv e to both prices and advertisement; and the third segment, with 13 percent, was sensitive to prices and advertising. All of the segments were separated by these factors as well as demographic characteristics of the households. He estimated the mean price elas ticity of demand for cut flowers and pot plants to be –0.28 for non-habitual b uyers, and –0.81 for habitual buyers. Behe (1989) in her study “Floral Purchase Behavior of Pennsylvanians” analyzed the consumer purchasing behavior in the fl ower industry at the retail level. She recommended three ways to segment retail flower markets based on buyers attitudes toward flowers, income and demographic ch aracteristics of the Pennsylvania households. The first type of segmentation was se gmentation by product, which basically distinguished between fresh cut flowers and potted plant customers. The second type of segmentation was segmentation by volume of pu rchase, where she had three categories: light, medium and heavy consumers. The thir d and final segmentati on was by location of the purchase, which included supermarkets or florists.

PAGE 58

38 Behe et al. (1992a), carried out an anal ysis of consumer pur chases of floral products in Ohio supermarkets. They used a mail-survey instrument to identify the factors that influence purchases of floral pr oducts in supermarkets. They used a principal components analysis and identified 34 indepe ndent factors that accounted for 64 percent of the variation. These factor s were grouped into five differe nt categories as follows: product, consumer, store, use (gifts vs. self) and use (location). These factors represented the most important influences on floral buyin g decisions and were us ed to define five different market segments of supermarket floral customers. In the same year, Behe et al. (1992b) carried out a follow-up study titled “Market Segmentation of Supermarket Floral Customer s.” They used the same data set obtained from the mail survey instrument and identified the 34 most important factors affecting floral buying decisions. They applied cluster an alysis on survey responses to create five homogeneous consumer segments. There were 14 factors that contributed to most of the differences among segments, including fact ors of product assortment, number of purchases, degree of personal use, and pack aging importance. The five clusters or segments were: (1) friendly buyers, which com posed 20 percent of to tal customers; (2) married man, which also had 20 percent of total customers; (3 ) Selfers, with 30 percent of the sample; (4) annual buyers, with 25 percen t of the total consumer sample; and (5) educated mothers with only 5 percent of the total sample. The importance of this study is the finding that clusters can be used by supe rmarkets and florists managers to target different potential market segments. Becker (1993) in his study “Products, Se rvices, and Consumer Perceptions of Service Quality in the Retail Floral Industry of Texas”, pres ented differences in service

PAGE 59

39 quality between supermarkets and florists in the Texas region. The differences in the types of retail outlets were ba sed in terms of types of products sold, custom design and other in-store services, deliv ery options and convenience. B ecker found that there were four main differences between florists and su permarkets. First, florists carried out more non-perishable items than supermarkets. Seco nd, florists provided more custom design compared to supermarkets. Third, almost all florists (99 percent) had delivery services compared to only 41 percent of supermarkets And fourth, the feat ure convenience was better perceived at supermarkets, because ofte n they open more days of the week and for longer hours. Ward (1997) evaluated PromoFlorÂ’s imp act on the demand for fresh cut-flowers and greens in the U.S. using household data on flower consumption. Ward suggested that changes in the flower demand occurred because either change of the number of buyers from period to period with the entry of new buyers that increase the number of buyers of floral products, or changes in purchases per buyer. Ward found that the impact of PromoFlor was different between the four income groups that he had defined: under $25,000; $25,000-$49,999; $50,000-$74,999; and $75, 000 or more. For the first three income groups that are for people with in come under $75,000, there were positive gains on sales with the PromoFlor campaign. House holds with incomes of $75,000 or more did not respond to the PromoFlor programs. Ward estimated that for every dollar spent on generic advertising at the handler level ther e was a gain of $5.6 dollars in additional revenue. Ward concluded that PromoFlor in creased by about 10 percent the number of households buying fresh cut-flow ers in a typical month.

PAGE 60

40 Rimal (1998) analyzed the e ffects of generic and brand promotions of fresh cutflowers in the use of retail flower outlets. Rimal used data on household purchases of fresh cut-flowers through all types of outlets to conduct an expenditure allocation model that used the Almost Ideal Demand Systems (A IDS) specification to estimate change in household expenditures on fresh cut-flower s among three flower outlets: florists, supermarkets and other outlets. Rimal grouped f actors that affect expenditure allocation among outlets in five categories: price effects, expenditure e ffects, advertising effects, seasonality effects, and behavioral effects. Rimal found that rela tive prices, income, seasonality, attitude and prom otion affected expenditure allocation on outlets for fresh cut-flowers. Rimal concluded that PromoF lor was generally market share neutral; however, it had significant and positive imp acts on florists in the income group $25,000$49,999. Overall the impact of brand adver tising on florists was positive, while for supermarkets it was negative. Girapunthong (2002) analyzed the demand dr ivers for fresh cut-flowers and their substitutes. In order to measure the factor s influencing the demand for flowers such as prices, seasonality, and demographic charact eristics, she estimated the demand for flowers in different forms, applying the Almost Ideal Demand Systems (AIDS) to examine household behaviors in the U.S. flower industry. Her model for fresh cut flowers, potted and dry/artifi cial flowers accounted for diffe rences in outlets, purposes, purchasing occasions, growth, income, demogr aphic characteristics of the households, and prices. Girapunthong found that all direct price effect coe fficients with the seasonal and actual variables were statistically signifi cant different from zero at the 95 percent confidence interval. Changes in the relative pr ices had a significant impact for flower

PAGE 61

41 market shares among fresh cut-flowers, potted fl owering plants and dry/ artificial flowers. The coefficient estimates for expenditure eff ects were significant, indicating that changes in total household expenditures for flowers ha d a significant effect on market shares. Ward (2004) evaluated the impact of a promotion program developed by the Flower Promotion Organization (FPO) that had the objective of increasing the frequency of buying of fresh cut flowers among existing female flower buyers in non-traditional holiday and non-event periods. To measure the impact of the FPO program, Ward estimated market penetration models and buyer frequency models to determine if the promotions had stimulated the demand for fr esh cut-flowers and then determined the value of the gains attributed to the pr omotion program. Ward concluded that the promotions have impacted the demand for flowers through increa sing buyer frequency and through attracting new buyers He found that abou t 87 percent of the increases in demand for the promotion programs are from th e increased transactions per buyer. Ward found that the demographic group that responde d the most to the promotion program was female buyers that purchases fl owers for self-use. This was c onsistent with the target of the FPO promotion program. All of the studies described above have made significant contributions to the flower industry in the Unite d States. From analyzing att itudes of consumer behavior toward floral products to esti mating the demand for flowers and analyzing the effects of both generic and brand advertisement programs, these studies have used both mail survey instruments and extensive household data in order to generater analyses. However, none of these studies have separated the demand fo r flowers in terms of market penetration models and buyer frequency models. Therefore, this research is expected to make

PAGE 62

42 substantial contributions in the field of economics and marketing by addressing this important topic.

PAGE 63

43 CHAPTER 3 UNITED STATES FLORAL INDUSTRY This chapter consists of descriptive statis tics for the flower industry in the United States. The chapter is organized in six sections. First, an introduction with specifications on general characteristics for the flower indus try and the categorization of the number of households on each of the divided regions will be presented. Second, expenditures by region, flower type, purpose of the purchase, gender, age, income and seasonal effects are presented and discussed. Third, number of transactions by flower type, purpose of the purchase, gender, age and income are addres sed. Fourth, expenditu res per transaction by flower type and demographics are discussed. Fifth, expe nditures per buyer or average price paid by buyers are computed by different demographic characteristics and by flower type. Sixth, market penetra tion by flower type, region, pur pose of the purchase, age, gender and income are presented and discussed. Introduction The data set for flower purchases from July 1992 to July 2004 was obtained from the American Floral Endowment (AFE) and Ipsos-NPD group. These data were based on a consumer panel of several thousand househol ds who reported their purchases of floral products in the US. The data set is organi zed by number of households, expenditures, transactions, and buyers. Each one of these categories contains data on demographic characteristics including region, purpose of the purchase, product form or flower type, gender, age and income. There are nine regions throughout this chapter analyzed in terms of differences on regions based on consumer be havior and demographi c characteristics of

PAGE 64

44 the population within each region. The nine regions are New England, Middle Atlantic, East North Central, West North Central, Sout h Atlantic, East South Central, West South Central, Mountain and Pacific. There are four basic product fo rms that are of particular interest on this study: fresh cut-flowers, flowering and gr een house, dried or artificial flowers, and out-door plants. Cut-flowers ar e subdivided into flower arrangements and non-arrangements. The purpose of the purchase is either for self-use or to use as gifts. Gender is either male or female. Age consiste d of four categories as follows: 25 years old or less, 25 to 39 years old, 40 to 54 years old, and 55 years old or more. Income is also divided into four categories as follo ws: less than $25,000, $25,000 $49,999, $50,00074,999, and $75,000 or more. Figure 3.1 Total Number of Households by Region from 1993 to 2003. Source: AFE and Ipsos-NPD group. The number of households increased 13.16 percent from 73.7 million in 1993 to 83.4 million in 2003. Regions with the greater number of households are South Atlantic, East North Central, Pacific a nd Middle Atlantic for a combin ed value of 64.7 percent of 19931994199519961997199819992000200120022003 0 20 40 60 80 100 Millions Number of Households New England 5.3% Middle Atlantic 14.6% East North Central 16.9% West North Central 7.2% South Atlantic 18.3% East South Central 6.2% West South Central 10.9% Mountain 5.7% Pacific 14.9%

PAGE 65

45 the total number of households. Figure 3.1 represents the total number of households from 1993-2003 by region and the aggregate level. Expenditures For Flowers Expenditures for flowers increased 82.16 pe rcent from 1993 to 2003. Of the total expenditures on flowers in the U.S., 18.3 per cent belongs to the South Atlantic region, 17.7 percent to East North Central and 16. 1 to Middle Atlantic. These three regions combined represent more than one half the to tal expenditures in the country. Figure 3.2 is a graphical representation of the e xpenditures in the U.S. by regions. Figure 3.2 Total Household Expenditure s on Flowers by Region from 1993 to 2003. Source: AFE and Ipsos-NPD group. During the 1993 to 2003 period the product fo rm with the highest expenditures were cut-flowers with a share of 36.3 per cent, followed by outdoor with 34.8 percent of the total expenditures. Figure 3.3 shows the distribution of expendi tures by flower type during the 1993 to 2003 period. Furthermore, fresh cut-flowers were subdivided into flower arrangements and non-arrangements. Mo st expenditure of cut-flowers was for flower arrangements with 54.5 percent of th e total expenditures on cu t-flowers, as shown in Figure 3.4. 19931994199519961997199819992000200120022003 0 1 2 3 4 5 6 7 Millions Expenditures New England 6.2% Middle Atlantic 16.1% East North Central 17.7% West North Central 6.7% South Atlantic 18.3% East South Central 5.9% West South Central 9.6% Mountain 5.0% Pacific 14.7%

PAGE 66

46 Figure 3.3 Total Household Expenditures on Flowers by Flower T ype, from 1993 to 2003. Source: AFE and Ipsos-NPD group. Figure 3.4 Total Household Expenditures on Flowers by flower type, Share of CutFlowers Expenditures from 1993 to 2003. Source: AFE and Ipsos-NPD group. Cut Flowers 36.3% Green House 19.7% Dry & Artificial 9.1% Outdoor 34.8% Arrangements 54.5% Non Arrangements 45.5% 1993-2003 2003 19931994199519961997199819992000200120022003 0.00 0.50 1.00 1.50 2.00 2.50 3.00 Expenditures in Millions Dry & ArtificialGreen House OutdoorCut Flowers Cut Flowers 36.3% Green House 19.7% Dry & Artificial 9.1% Outdoor 34.8% Cut Flowers 38.6% Green House 19.3% Dry & Artificial 5.4% Outdoor 36.7%

PAGE 67

47 The purpose of the purchase of floral products was either for self-use or to use as gifts. During the 1993 to 2003 period, consumer s spent about the same amount for selfuse and for gifts including all fl ower types, all regions and all households. About one half of the purchases were for self-use and one half for gifts. This relationship is represented graphically in Figure 3.5. Figure 3.5 Total Household Expenditures on Flowers by Purpose from 1993 to 2003. Source AFE and Ipsos-NPD group. In terms of gender, females are the group with the largest sh are of expenditures with 77.3 percent from the 1993 to 2003 period. Figure 3.6 represents total expenditures by gender from 1993 to 2003. There is a dir ect relationship between the age of the households and the expenditures; the younger th e household category, the less it spends on flowers. This relationship is shown in Figure 3.7. The ag e group with the largest share of household expenditures is the oldest age gr oup (55 years of age or more), with 40.9 percent of total house hold expenditures. 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Expenditures in Millions SelfGift Self 49.8% Gift 50.2% Self 50.8% Gift 49.2%

PAGE 68

48 Figure 3.6 Total Household Expenditure s on Flowers by Gender from 1993 to 2003. Source: AFE and Ipsos-NPD group. Figure 3.7 Total Household Expenditures on Flowers by Age Groups from 1993 to 2003. Source: AFE and Ipsos-NPD group. 1993-2003 2003 19931994199519961997199819992000200120022003 0.00 1.00 2.00 3.00 4.00 5.00 6.00 Expenditures in Millions Male Female Male 22.7% Female 77.3% Male 24.4% Female 75.6% 1993-2003 2003 19931994199519961997199819992000200120022003 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Expenditures in Millions under 25 yrs25-39 yrs 40-55 yrsover 55 yrs Under 25 4.4% 25-39 22.4% 40-55 32.3% over 55 40.9% Under 25 3.5% 25-39 18.9% 40-55 33.0% over 55 44.6%

PAGE 69

49 The income group with the la rgest share of ex penditures is the highest income group ($75,000 or more), with 41 percent of the household expenditures. The second income group ($25,000 to $49,999) was ranked second in terms of household expenditures with a share of 24 percent of the householdÂ’ s expenditures. The lowest income group spends the least on flowers, followed by the third income group. These results suggest that the higher the income th e more expenditure on fl owers, except for the third income group ($50,000 to $74,000) wher e we observe a truncation in expenditure patterns. This behavior is summarized in Figure 3.8. Figure 3.8 Total Household Expenditures on Flowers by Income Gr oups from 1993 to 2003. Source: AFE and Ipsos-NPD group. Figure 3.9 shows seasonal pattern expe nditures on flowers by flower type. As shown in the graph, for cut-fl owers there is an increase in expenditures during February, and May, which correspond to ValentineÂ’s Da y and MotherÂ’s Day. Greenhouse, and dry and artificial flower expenditures increased during the April-May period and December, 1993-2003 2003 19931994199519961997199819992000200120022003 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 Expenditures in Millions Under $25,000$50,000-$74,999 $25,000-$49,999Over $75,000 Under $25,000 16.8% $25,000-$49,999 24.0% $50,000-$74,999 18.1% Over $75,000 41.0% Under $25,000 12.6% $25,000-$49,999 21.5% $50,000-$74,999 18.1% Over $75,000 47.8%

PAGE 70

50 which corresponds to MotherÂ’s Day and Christmas. Outdoor plants have a substantial increase during the months of April, May a nd June with a peak of 37 percent of the expenditures on outdoor plants that occur in May. Figure 3.9 Shares of Seasonal Expenditures on Flowers by Flower Type during the 1993 to 2003 Period. Source: AFE and Ipsos-NPD group. Transactions on Flowers The number of transactions is reported on a monthly basis. Differing from regular consumer demand studies that employ the total quantity consumed on a given period, this study uses the number of transactions per month as a substitute for the quantity. The reason for using the number of transactions instead of the total quantity consumed on flowers is simple. For a given period, a consumer may buy a flower arrangement consisting of a dozen flowers for example, or that consumer may only buy one flower. There is no way for the resear cher to know the exact quantity of flowers used on different JanFebMarAprMayJunJulAugSepOctNovDec 0.00 0.10 0.20 0.30 0.40 Share of Expenditures Cut FlowersGreen House Dry/ArtificialOutdoor Cut Flowers0.070.140.080.090.130.070.070.060.070.070.070.08 Green House0.050.070.080.130.150.070.050.050.050.050.060.17 Dry/Artificial0.070.090.100.100.120.070.070.070.080.070.080.09 Outdoor0.010.020.070.180.370.140.050.030.050.040.020.01

PAGE 71

51 flower products. However, the data utilized in this study are aggreg ate data of consumer expenditures for flowers, and the illustration of a single household be havior is used for exemplification purposes. An alternative approach is to substitute the transactions for the quantities. The number of tran sactions by month and by flow er type during the 1993 to 2003 period is presented in Figure 3.10. Figure 3.10 Number of Tran sactions by Month and by Flow er Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD group. The number of transactions is higher for consumers who bought flower products for self-use, as shown in Fi gure 3.11. This in dicates that consumers who buy floral products for themselves do it more freque ntly over a given period. The number of transactions per buyer over a monthly peri od has been relatively stable with low fluctuations from 1993 to 2003. 19931994199519961997199819992000200120022003 2.50 2.60 2.70 2.80 2.90 3.00 3.10 3.20 3.30 3.40 3.50 3.60 3.70 3.80 Number of Transactions Cut FlowersGreen House Dry/ArtificialOutdoor Cut Flowers2.732.792.712.852.732.772.793.022.792.792.83 Green House2.602.712.582.672.632.782.722.662.652.692.65 Dry/Artificial3.163.213.423.563.483.523.613.543.393.463.15 Outdoor3.343.363.483.503.413.603.563.763.673.713.59

PAGE 72

52 Figure 3.11 Number of Transa ctions by Purpose During the 1993 to 2003 period. Source: AFE and Ipsos-NPD group. Figure 3.12 Number of Tran sactions by Gender During the 1993 to 2003 period. Source: AFE and Ipsos-NPD group. 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 0.50 1.00 1.50 2.00 2.50 Number of Transactions GiftSelf Self 58.2% Gift 41.8% Self 58.2% Gift 41.8% 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 0.50 1.00 1.50 2.00 Number of Transactions MaleFemale Male 45.8% Female 54.2% Male 47.2% Female 52.8%

PAGE 73

53 This result is expected, sin ce the majority of consumers who buy product for gifts do so mostly on calendar occasions. In terms of ge nder, women complete more transactions during a given period as compared to men. Figure 3.12 represents the number of transactions by gender. The number of tran sactions by age, for the three higher age groups, has not changed considerably from 1993 to 2003. The onl y group with high fluctuations on the number of transacti ons is the younger group. Figure 3.13 is a graphical representation of the number of transactions by age. At this point it is not clear what happened in 1995among the transacti ons for the youngest age group, households under 25 years of age. Figure 3.13 Number of transa ctions by Age Groups for th e 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. The income group with the la rgest number of transacti ons is the hi ghest income group ($75,000 or more). As shown in Figure 3. 14, there has been an increase in the 19931994199519961997199819992000200120022003 1.00 1.10 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 Number of Transactions under 25 yrs25-39 yrs 40-55 yrsover 55 yrs under 25 yrs1.821.551.001.451.811.641.951.561.741.581.81 25-39 yrs1.591.491.601.631.641.701.571.661.701.701.69 40-55 yrs1.581.681.551.671.651.731.761.721.661.681.68 over 55 yrs1.561.601.631.661.591.591.641.711.661.641.72

PAGE 74

54 number of transactions for all four-income groups from 1993 to 2003. The lowest income group, consisting of households with annual in come of $25,000 or less, had an increase of about 12 percent in the number of transactions to 1. 69 in 2003, up from 1.51 in 1993. The number of transactions by income for 2003 is very similar for the four income groups, differing with less than 3.6 percent from the lowest number of transactions (second income group) to the highest number of transactions (hi ghest income group). Figure 3.14 Number of Transactions by In come Groups for the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.15 represents seasonality of th e number of transactions per month on flowers by flower type. The seasonality for cut-flowers and greenhouses has been relatively stable. There have not been dram atic fluctuations in any given month as represented on the relatively flat curves on th e graph. For outdoor plants there is a peak of 1.96 during May, while dry and artificial plants have tw o peaks in March and August. The number of transactions by region is very similar for all regions as represented in 19931994199519961997199819992000200120022003 1.50 1.55 1.60 1.65 1.70 1.75 1.80 1.85 Number of Transactions Under $25,000$50,000-$74,999 $25,000-$49,999Over $75,000 Under $25,0001.511.561.641.641.701.661.631.661.611.681.69 $50,000-$74,9991.571.511.541.551.601.701.661.701.711.621.67 $25,000-$49,9991.631.621.521.681.531.581.691.651.611.701.72 Over $75,0001.641.711.661.771.691.711.751.801.741.701.73

PAGE 75

55 Figure 3.16. The regions with the highest numbe r of transactions are East South Central and West South Central. Figure 3.15 Seasonality of the Number of Transactions per Month for the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.16 Number of Tran sactions by Region from the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. JanFebMarAprMayJunJulAugSepOctNovDec 1.20 1.30 1.40 1.50 1.60 1.70 1.80 1.90 2.00 2.10 Number of Transactions Cut FlowersGreen House Dry/ArtificialOutdoor Cut Flowers1.321.341.371.331.351.371.371.371.371.321.361.39 Green House1.321.281.301.321.311.291.281.251.231.271.291.30 Dry/Artificial1.601.671.811.671.731.671.591.811.661.621.671.53 Outdoor1.441.551.721.871.961.811.651.571.591.611.581.38 1.79 1.72 1.75 1.79 1.75 1.84 1.83 1.74 1.79 New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 2.50 Number of Transactions

PAGE 76

56 Expenditures per Transaction Expenditures per transacti on are the average amounts of money consumers spend per transaction. Expressed in a different manne r is the average price per transaction. This is of vital importance, since number of transactions and the average price per transaction are two key variables for measuring the c onsumer demand for flowers. Among the four flower types, consumers spend the most pe r transaction on cut-flowers as shown in Figure 3.17. Also, consumers spend more per transaction when the purpose of the purchase was for gifts. This is represented in Figure 3.18. Figure 3.17 Expenditures per transaction by Flower Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.19 shows expenditures per tran saction by gender with males spending more per transaction than females. In 2003, males spent 32.24 percent more per transaction than females. The average expe nditure per transaction for males in 2003 was $17.68, compared to $13.37 for females. 19931994199519961997199819992000200120022003 15.00 20.00 25.00 30.00 35.00 Expenditures per Transaction $ OutdoorGreen House Dry & ArtificialCutFlowers Outdoor19.3221.6316.2217.5317.9018.6119.7819.6418.9821.6221.20 Green House20.0320.9921.0019.5021.9421.1122.7923.1421.7222.8726.07 Dry & Artificial21.0118.9719.4619.8719.4023.1121.5823.9028.2724.4228.01 CutFlowers24.1926.5827.0826.9932.7232.0632.5732.6230.5829.3329.89

PAGE 77

57 Figure 3.18 Expenditures per Transaction by Purpose During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.19 Expenditures pe r Transaction by Gender During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 20030.00 5.00 10.00 15.00 20.00 Expenditures per Transaction $ SelfGift Self 33.4% Gift 66.6% Self 34.5% Gift 65.5% 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 5.00 10.00 15.00 20.00 Expenditures per Transaction $ FemaleMale Male 54.7% Female 45.3% Male 56.9% Female 43.1%

PAGE 78

58 Consumers of 25 years of age or less ha ve the lowest expenditures per transaction with an average of $9.4 for 2003. Th e age group with highe st expenditures per transaction is the category of consumers betw een 40 and 55 years of age with an average expenditure per transa ction of $14.67 for the year 2003. Figure 3.20 represents the average expenditure per transaction by age groups. In terms of income, the highest income group ($75,000 or more) spent the most per transaction, fo llowed by the third income category ($50,000 to $74,999), with average expend itures per transaction of $16.15 and $14.45, respectively. The lowest income group ( $25,000 or less) showed the least expenditures per transact ion with an average of $11.25 in 2003. Figure 3.21 shows the average expenditure per transaction for the four income groups during the 1993 to 2003 period. Figure 3.20 Average Expend itures per Transaction by Ag e Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 19931994199519961997199819992000200120022003 7.00 8.00 9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 Expenditures per Transaction $ under 25 yrs25-39 yrs 40-55 yrsover 55 yrs under 25 yrs7.6010.378.528.5410.5810.4411.759.6812.3010.869.40 25-39 yrs10.9413.9512.4013.9914.1413.3313.6914.2713.6815.3614.03 40-55 yrs13.4313.4415.0614.0215.2114.5914.7616.0315.0414.6414.67 over 55 yrs11.3612.5711.5411.8112.5115.0013.7112.2413.2613.8313.06

PAGE 79

59 Figure 3.21 Average Expenditures per Trans action by Income Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.22 Average Expenditure per Tr ansaction by Region during the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 19931994199519961997199819992000200120022003 8.00 10.00 12.00 14.00 16.00 18.00 20.00 Expenditures per Transaction $ Under $25,000$25,000-$49,999 $50,000-$74,999Over $75,000 Under $25,0009.2910.499.8110.1811.3412.4211.0012.1313.3411.7911.25 $25,000-$49,99911.3013.2512.6513.3312.5212.6413.2712.8912.2613.5113.27 $50,000-$74,99914.1916.4215.8014.4417.0417.3715.6416.7414.6414.6814.45 Over $75,00013.8112.8714.7414.5316.4216.2217.2516.6515.9316.9216.15 11.25 12.23 11.77 11.53 12.27 13.70 12.58 11.99 10.96 New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 5.00 10.00 15.00 20.00 Expenditures per Transaction $

PAGE 80

60 Figure 3.22 is a graphical representation of the average expenditure per region. The region with highest expenditures per transaction is East South Central, with an average of $13.70 during the 1993 to 2003 period. The Pacifi c region is the region with the lowest average expenditure per tran saction with an average of $10.96 per transaction. Expenditures Per Buyer Expenditures per buyer are the averag e amount by each buyer depending on the number of transactions that specific buyers had during a m onthly period. This measure will help to describe even further consum er attitudes toward floral products. Average price paid by consumers was obtained by divi ding total expenditures by buyers ($/buyer). The product form with the highest expenditu res per buyer is cut flowers with $40.33 in 2003, up from $31.15 in 1993; this represents almost a 30 percent increase during that time period. Figure 3.23 is the graphical repr esentation of average expenditures per buyer from 1993 to 2003. Figure 3.23 Average Expenditures by Buye rs by Flower Type During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 19931994199519961997199819992000200120022003 20.00 25.00 30.00 35.00 40.00 45.00 50.00 Expenditures per Buyer $ CutFlowersGreen House Dry & ArtificialOutdoor CutFlowers31.1534.9934.9136.7242.3842.0543.3445.7741.3239.4540.33 Green House24.9026.9326.1425.3628.3127.5129.6729.4027.5629.5133.73 Dry & Artificial29.4529.2629.7030.1031.6636.4035.1035.5141.7034.5439.33 Outdoor30.3333.2427.6029.6530.0832.3634.9134.7633.4637.4836.36

PAGE 81

61 Figure 3.24 and 3.25 show the average pr ice paid by buyers by purpose and gender, respectively. Figure 3.24 Average Expenditures by Buye rs by Purpose During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.25 Average Expenditures by Buye rs by Gender During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 20030.00 5.00 10.00 15.00 20.00 25.00 30.00 Expenditures per Buyer $ SelfGift Self 41.6% Gift 58.4% Self 43.0% Gift 57.0% 1993-2003 2003 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 0.00 5.00 10.00 15.00 20.00 25.00 30.00 Expenditures per Buyer $ MaleFemale Male 51.6% Female 48.4% Male 55.5% Female 44.5%

PAGE 82

62 The 25 years of age or less age group is the category with the lowest average expenditures per buyer with an average of $15.82 in 2003, up from $12.64 in 1993, which represents a 25 percent increase. The age group between 40 and 55 years of age has the highest average expe nditure per buyer, with an average of $23.47 in 2003, up from $20.54 in 1993. The lowest age category has some changes over time; in contrast the other three categories have maintained a trend throughout the pe riod with few abrupt changes as shown in Figure 3.26. Figure 3.26 Average Expenditures by Buye rs by Age Group During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. In terms of income, households with overa ll income of $75,000 or more show the most expenditures per buyer, followed by the $50,000 to $74,999 group with average expenditures per transaction of $26.64 and $23. 39, respectively. Households with income of $25,000 or less have the lowest expenditu res per buyer with an average of $18.19 in 19931994199519961997199819992000200120022003 5.00 10.00 15.00 20.00 25.00 30.00 Expenditures per Buyer $ under 25 yrs25-39 yrs 40-55 yrsover 55 yrs under 25 yrs12.6416.158.5212.4816.4815.2822.4714.3719.4317.3115.82 25-39 yrs16.5420.2718.7021.6222.4421.1621.1122.6121.8524.3222.39 40-55 yrs20.5421.7522.3022.1223.8823.9024.5425.7423.7023.6423.47 over 55 yrs16.7218.9217.7518.1918.7422.6021.0120.1220.5421.4421.30

PAGE 83

63 2003. Figure 3.27 shows the average expenditu re per buyer for the four income groups during the 1993 to 2003 period. For all inco me groups, except for the $50,000 to $74,999 income category, there has been an increa sing trend in expenditures per buyer from 1993 to 2003, with some variations within the period. Figure 3.27 Average Expenditures by Buyers by Income Groups During the 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. The region with the highest average price paid by buyers is the East South Central with an average of $21.83, fo llowed by the West South Centra l region with an average of $20.59. Mountain is the region with the repor ted lowest expenditures per buyer among all regions. Differences in expend itures per buyer by region ar e very small as shown in Figure 3.28. 19931994199519961997199819992000200120022003 10.00 15.00 20.00 25.00 30.00 Expenditures per Buyer $ Under $25,000$25,000-$49,999 $50,000-$74,999Over $75,000 Under $25,00013.4515.7815.0616.1418.5319.4117.6419.2220.1019.2118.19 $25,000-$49,99916.5619.1218.5919.3818.8619.8520.5220.5019.7020.6620.75 $50,000-$74,99922.0725.6323.1722.3625.6026.4224.5826.6122.3623.5323.39 Over $75,00021.6021.0123.0824.0725.9526.5428.5827.7825.7227.0526.64

PAGE 84

64 Figure 3.28 Average Expenditures by Buye rs by Region During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Market Penetration Market penetration is define d as the number of buyers divided by the total number of households, varying between zero and one on the percentage of households that are buyers. The market penetration measure is vi tal for this study sinc e one of its main objectives is to separate the demand due to market penetratio n from that of frequency of buying. Flower types with the highest market penetration are outd oor plants and cutflowers as shown in Figure 3.29. Market pe netration is higher wh en the purpose of the purchase is for self-use as represented in Figure 3.30. 18.69 19.71 19.29 18.90 19.42 21.83 20.59 18.50 17.81 New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 5.00 10.00 15.00 20.00 25.00 30.00 Expenditures per Buyer $

PAGE 85

65 Figure 3.29 Percent of Market Penetration by Flower Type During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.30 Percent of Market Penetra tion by Purpose During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 19931994199519961997199819992000200120022003 0.00 0.05 0.10 0.15 0.20 0.25 0.30 Market Penetration % Cut FlowersGreen House Dry/ArtificialOutdoor Cut Flowers0.190.170.160.180.180.180.170.170.210.220.27 Green House0.170.160.160.170.170.160.160.160.170.170.20 Dry/Artificial0.080.080.080.090.070.070.060.060.060.050.05 Outdoor0.190.190.190.210.210.210.200.200.220.210.25 0.32 0.31 0.31 0.34 0.33 0.32 0.31 0.31 0.34 0.33 0.40 0.27 0.25 0.24 0.26 0.25 0.25 0.23 0.23 0.25 0.26 0.31 19931994199519961997199819992000200120022003 0.20 0.25 0.30 0.35 0.40 Market Penetration % GIFTSELF

PAGE 86

66 Market penetration values are higher for females compared to males as shown in Figure 3.31. Age groups have a direct relations hip with market penetration values, where the higher the age group, the higher the level of market penetration, as shown in Figure 3.32. Figure 3.31 Percentage Market Penetr ation by Gender During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. In Figure 3.33 the percentage values for market penetration by income shows households with $75,000 or more (highest in come group) to have the highest market penetration values. Surprising ly, the third income group (households with income of $50,000-$74,999) has the lowest market penetrat ion percentage values among the income groups. 0.47 0.45 0.43 0.48 0.47 0.45 0.43 0.43 0.48 0.48 0.56 0.12 0.11 0.11 0.12 0.12 0.11 0.11 0.11 0.12 0.12 0.15 19931994199519961997199819992000200120022003 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 0.45 0.50 0.55 0.60 Market Penetration % MALEFEMALE

PAGE 87

67 Figure 3.32 Percentage of Market Pene tration by Age Groups During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. Figure 3.33 Percentage of Market Penetra tion by Income Groups During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 19931994199519961997199819992000200120022003 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 Market Penetration % under 25 yrs25-39 yrs 40-55 yrsover 55 yrs under 25 yrs0.060.070.060.070.060.060.060.060.060.070.07 25-39 yrs0.330.300.280.310.280.250.210.210.210.230.27 40-55 yrs0.310.300.300.340.360.340.310.320.350.360.42 over 55 yrs0.480.460.440.490.470.480.500.500.570.540.66 19931994199519961997199819992000200120022003 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 Market Penetration % Under $25,000$25,000-$49,999 $50,000-$74,999Over $75,000 Under $25,0000.300.290.270.310.280.260.230.220.210.200.24 $25,000-$49,9990.360.330.300.300.330.300.270.270.290.290.33 $50,000-$74,9990.190.170.170.190.170.190.170.180.200.210.25 Over $75,0000.330.340.340.410.390.380.400.410.490.500.60

PAGE 88

68 Finally, market penetration among regions is similar ranging fr om 26 percent to 35 percent in the New England region. Note that in relative terms the South shows the lowest market penetration leve ls as shown in Figure 3.34. Figure 3.34 Percentage of Market Penetration Index by Region During 1993 to 2003 Period. Source: AFE and Ipsos-NPD Group. 0.35 0.31 0.30 0.28 0.30 0.260.260.26 0.31 New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 0.10 0.20 0.30 0.40 0.50 Market Penetration %

PAGE 89

69 CHAPTER 4 CONCEPTUAL FRAMEWORK AND THEORETICAL MODELS In this chapter the conceptual framework to explain consumer demand for flowers is presented. The main difference of this st udy from traditional demand analyses is that this research separates total demand into tw o components, that attributed to market penetration and to changes in buyer freque ncy. In order to present the conceptual framework and theoretical models, this ch apter has five sections. Consumer demand theory adapted specifically fo r the flower case is first pres ented, followed by a discussion of the nature of the data. This section show s why the Tobit model is the most appropriate for the analysis. The conceptual foundations of the Tobit model will be derived and explained, and then a market penetration mode l is constructed. Fina lly, a buyer frequency model is developed. Consumer Demand Theory for the Case of Flowers Consumer demand theory explains consumer preferences or consumer satisfaction obtained from the physical cons umption or use of goods and se rvices. This sa tisfaction or preference can be measured by the utility function and its properties, presented in an earlier chapter. The level of satisfaction can be described as a result of the consumption or use of a bundle of goods and services. Rational consumers purchase the optimal quantities of goods and services by maximizi ng the utility function subject to a budget constraint, as follows:

PAGE 90

70 (4.1) Max nq q u u ,...,1 S to: n j j jm q p1, where jp and jq are the price and the quantity of the jth good, respectively, and m is the total income. It is assumed that the income is equal to the sum of the expenditures on all goods; hence, there are no savings or total income is spent. Traditional demand analyses are based on the consumption of q, the total quantity consumed. As mentioned previously, the units for the quantity of flower s are not as straightforward as seen in many commodities. For example, a person may buy an arrangement of flowers, or a single rose. The two products can yield to that person similar levels of satisfaction (utility), but the units for the quantity are different. An alternative approach used to avoid this problem consists of using the number of transactions in a given period instead of the quantity. If one can obtain the number of buyers and the number of tran sactions for a given period, then the units in the model can be expresse d as the total frequency of purchasing or number of buyers times transacti on per buyer (Girapunthong, 2002). Let b represent the number of buyers and f represent the frequency of buying, then the income constraint on (4.1) can be modified to obtain: (4.2) n j n j j j j jm m b f p11, where jp represents the price per transaction. The Lagrangian procedure can now be used on the utility function, and then the consumerÂ’s demand function becomes: (4.3) j i i ip p m f b f ,

PAGE 91

71 The demand function now descri bes the consumption of goods in terms of a particular income, prices of the jth goods in quantity terms and pri ces of the ith good in transaction units or dollars per transacti on. This would be the utility function maximized in order to obtain the demand for flowers. An Overview of NPD Data Set The data set used in the models was obtained from the Ipso s and NPD group; the relevant variables used in this study are desc ribed on Chapter 3. In order to separate the total effect of the demand for flowers in th e U.S. into market penetration demand and buyer frequency demand, market penetration and buyer frequency models are developed. For each model two approaches will be used First regional and product form changes are allowed to have different intercepts and slopes. This means that there will be a different equation for each region and product form. Hence, there will be i j equations, where i is the number of product forms and j is the number of regions. The second approach will allow changes in the intercept and slopes produ ct forms, while regi onal changes are only in the slopes. The intercept of the regional changes, as well as the other dummy variables employed, will represent the average of all regions. This approach will have i equations, or one equation for each flower product fo rm, with the forms being cut-flowers, flowering plants, dry/ar tificial and out-door. For the first approach, the dependent variables for each of the models will be penetration and frequency. Pe netration is easily calculated by dividing the number of buyers by the number of househol ds as in equation (4.4): (4.4) ij ij ijHWD BUY PEN 1 0 ijPEN

PAGE 92

72 where ijPEN ijBUY and ijHWD are penetration, buyers and households for ith product form and the jth region. Frequency is derived by di viding transactions by buyers as in equation (4.5): (4.5a) ij ij ijBUY TRN FQ where ijFQ ijTRN and ijBUY are frequency, transactions and buyers for the ith product form and the jth region. By definition a person who is a buyer had at least one transaction or more in a given period, or else that person w ould not be defined as a buyer. Since ijFQ is censored at 1, an often-used option for estimation purposes is to adjust the censored variable so that the lower limit is zero. That adjustment simply entails subtracting the lower level from the original censored value. That is (4.5b) 1 ij ijFQ FRQ where ijFRQ is the adjusted frequency. The independent variables for both penetr ation and frequency models will be discrete variables created for income, gender, purpose, age and seasonal monthly dummies. If we employ the common method of creating du mmy variables described by Greene (2000), then the base level for all the coefficients of the dummy variables will be the category we left out of the equations in order the evade the dummy variable trap. A different approach consists of restricti ng the sum of the coe fficient of the dummy variables to zero. In this case, the base of the dummies would be the mean of all the categories. For example, let ki be the parameter estimate for income, then if the restriction 4 10k ki is imposed, then 4 2 1 k ki i is obtained and then the dummy

PAGE 93

73 variable 1inc inc dinck k will be created, where 1 k. More generally we would impose the restriction as follows: (4.6) K k ki 10 to obtain K k ki i 2 1 and in order to create the dummy va riables the following operation follows: (4.7) 1category category dummyk k where 1 k. The price per transaction is calculated from the data set by dividing expenditures by the transactions: (4.8) ij ij ijTRN EXP PRT where ijPRT ijEXP and ijTRN are price per transaction, expenditures and transactions for the ith product form and the jth region. A pr oblem arises for the penetration model; if a household was not a consumer by definition the number of transactions for that household in that period was zer o, and then the price would not be defined and in the data set would be considered zero. This would resu lt in an underestimation of the price, since the zeros would lower the demand effect fo r the market penetration model. Therefore price per transaction will only be included in the frequency models where purchases occurred. The second approach is very similar. The only difference is that instead of allowing the intercept to change for each region, that is calculating an equation for each region, dummy variables for regions will be created using the process described above; and then these dummy variables for the different regions will be added as independent variables to both penetration and frequency models.

PAGE 94

74 When a data set has a dependent variable that is zero for a significant fraction of the observations, conventional regression methods fail to account for the qualitative difference between limit (zero) observati ons and non-limit (conti nuous) observations (Greene, 2000). Tobit models have been used in single commodity studies to account for the fact that not all households purchase that co mmodity. Greene (2000) showed that in a regression model where a large proportion of the observations of the dependent variable are zero, then the ordinary least squares (O LS) parameter estimates tend to be biased toward zero, and the degree of bias depends on the amount of censoring (Cornick et al., 1994). For these type of data the most appropr iate model is the one developed by Tobin, the Censored Regression Model or simply the Tobit model. Tables 4.1 and 4.2 show the percentage of observations for which the dependent variable for penetration models are zero, and the percentage of the observations for which the dependent variable for the frequency mode ls are one. Note that the frequency table show the percentage of households that are in the censored limit of one, since it is more meaningful to explain in the table as the households who made only one transaction in a given period, and hence became buyers. Table 4.1 Percentage of Observations of Pe netration Model Dependent Variable That Are Censored at Zero. Source: AFE and Ipsos Group. % of pen=0ALLNEMAENCWNCSAESCWSCMOUNTPACIF TOTAL5.87%44.76%24.72%26.07%45.01%24.44%49.64%36.33%47.41%25.63% indoor8.08%51.07%30.25%31.30%51.09%31.12%56.42%43.54%54.98%32.52% cutflowers19.28%63.53%43.02%45.33%67.15%46.73%76.70%61.47%72.45%46.80% Arrange48.09%89.77%75.60%75.18%85.95%75.46%89.10%83.56%88.43%79.32% Non Arran20.61%67.58%47.08%50.44%73.07%53.05%83.45%68.07%78.07%51.15% Green House18.28%72.23%54.17%51.84%70.99%52.21%75.16%63.70%73.61%51.96% Flower27.68%78.90%63.16%61.87%78.84%61.88%82.26%73.88%81.46%62.79% Foliage33.58%87.46%73.69%70.55%84.05%70.21%87.07%77.92%86.44%68.95% Dry & Artificial44.12%90.99%76.22%71.54%81.39%70.08%81.58%77.16%87.93%80.19% Outdoor26.95%76.45%63.16%63.94%76.50%55.23%77.47%68.55%77.96%55.87% TOTAL5.87%44.76%24.72%26.07%45.01%24.44%49.64%36.33%47.41%25.63%

PAGE 95

75 Table 4.2 Percentage of Observations of Fr equency Model Dependent Variable That Are Censored at One. Source: AFE and Ipsos Group. In our household data set a la rge proportion of the dependen t variables for both models are censored at zero or one. Therefore, proced ures that take into consideration censored expenditures distributions must be used (i.e., the Tobit model). The Tobit Model Nobel prize winner Tobin (1958), showed th at when the dependent variable in a regression model equation has a lower or upper limit, and th e dependent variable takes the value of the limit for a large number of sample observations, conventional multiple regression analysis is not an appropriate technique to be used (Lung-Fei and Maddala, 1985). As shown in Tables 4.1 and 4.2, a large proportion of the penetration and frequency data take the value of the limit. In order to account for this truncation on the data set, Tobin developed a model specified as follows: (4.9) i i ix y *, where ix is a (1 K) vector of explanatory variables and ) 0 ( ~2 Ni and it is independent of other errors. The problem arises because in order for a household to be a buyer, it has to have at least one transa ction during a given period. Adjusting the % of freq=1ALLNEMAENCWNCSAESCWSCMOUNTPACIF TOTAL13.13%45.81%34.56%34.95%45.19%32.81%46.20%41.12%49.35%34.17% indoor19.19%55.18%41.47%41.86%54.41%41.95%55.60%49.22%60.35%43.56% cutflowers28.69%64.64%54.41%57.74%70.53%58.73%75.83%63.73%72.51%59.19% Arrange66.35%86.64%80.35%83.73%85.02%83.07%86.77%83.10%87.34%85.31% Non Arran33.20%66.10%58.21%60.56%73.85%61.15%75.93%65.31%73.13%61.22% Green House39.64%69.05%60.32%60.14%70.83%60.40%69.81%66.01%71.83%59.27% Flower49.06%74.91%66.95%68.24%76.87%68.00%76.13%73.20%78.70%67.72% Foliage55.94%76.38%74.89%69.71%77.76%72.35%77.77%74.94%75.28%69.60% Dry & Artificial35.34%62.05%54.88%50.13%54.29%51.80%56.36%54.20%64.93%55.20% Outdoor29.14%45.39%43.51%40.90%39.61%40.43%44.70%43.93%45.99%40.64% TOTAL13.13%45.81%34.56%34.95%45.19%32.81%46.20%41.12%49.35%34.17%

PAGE 96

76 subtracted one from the frequency variable to have the lower limit equal zero. In the penetration model a large number of the observations take the value of the lower limit, zero. Thus for any household the penetration and frequency models would take the form: (4.10) i iy y if 0*iy 0 iy if 0*iy. From the total number of observations T in the sample, the number of observations can be divided into tw o groups; one for which 0 iy 0T; and another for the number of observations for which 0 iy, 1T. In order to observe the statistical problems arising from the censored sample problem, consid er leaving out of the analysis the 0T observations for which 0 iy. For the remaining 1T sample observations, they are complete observations. Hence, one can use least squares estimators to estimate The problem is that this estimator is biased and inconsistent. In order to prove that, one can write down the expectation of the observed values of iy conditional on the fact that 0 iy: (4.11) 0 | 0 | i i i i iy E x y y E If the conditional expectation of the error term is zero, th en the estimates of the least square regression on 1T would provide an unb iased estimator for However this is not the case; if the i are independent and normally distri buted random variables, then the expectation would be: (4.12) 0 | 0 | i i i i ix E y E. It can be shown that this conditional expectat ion can also be expressed in the following manner:

PAGE 97

77 (4.13) i i i i ix E |, where i and i are the standard normal probability distribution function (p.d.f), and cumulative distribution function (c.d.f.) evaluated at ) / ( ix ; therefore in the regression model, if 0 iy, then, (4.14) i i i i i i iu x x y if we apply the regular least squares procedures the term i i is omitted. Since that term is not independent of ix the results are biased and inconsistent. In order to estimate the parameters and 2 consistently, maximum likelihood estimation (MLE) procedures can be used. The likelihood function of the sample has a component for the observations that are positiv e, and one for the observations that are zero. For the observations 0 iy it is known that 0 i ix or expressed in a different way, i ix then, (4.15) i i i i i ix x y 1 Pr Pr 0 Pr If we define the product of the observations over the zero lower limit level to be 0 and the product over the positive observations to be 1 the likelihood function of the Tobit model is given by: (4.16) 2 2 2 1 2 1 02 exp 2 1 i i ix y The corresponding log-like lihood function would be:

PAGE 98

78 (4.17) 2 2 1 2 1 1 02 ln 2 / ) 2 ln( 2 / 1 ln ln i i ix y T T L Then the first order conditions are: (4.18a) i i i i ix x y i x L 1 2 01 1 1 (4.18b) 2 1 4 2 1 0 3 22 1 2 1 ) ( 2 1 i i i i ix y T x L The Time Series Processor (TSP) has a routine to maximize the log likelihood function for the Tobit model. The Tobit rou tine uses the analytic first and second derivatives to obtain maximum likelihood es timates via the Newton-Raphson algorithm. The starting values for the parameters are ob tained from a regressi on on the observations with positive y values. The numerical impl ementation involves evaluating the normal density and cumulative normal distributi on functions. The cumulative distribution function is computed from an asymptotic expansion, since it has no closed form. The ratio of the density to the dist ribution function, used in the derivatives, is also known as the Inverse Mills Ratio (Hall, 1992). Market Penetration Models There are two approaches used to develop the market penetration models. The first approach, Penetration Model I, uses a T obit model equation for each product form i and marketing region j In consequence there will be one equation for each product form and region, for a total of i j equations for the first approach. This approaches captures changes in both the intercept and the slope of the indepe ndent variables for market penetration. The second approach, Penetrat ion Model II, incor porates regions as

PAGE 99

79 independent dummy variables. Therefore th e regions would have a common slope but different intercepts depending on the regional dummy estimates. Penetration Model I Let iX represent income, gender, purpose, ag e, and seasonal monthly expenditures. Then the general Model I would be expressed with any reference to the observation being dropped for notational convenience as follows: (4.19) ij i ij iju X PEN PEN if 0*ijPEN 0ijPEN if 0*ijPEN, which is the classical Tobit model described in the previous section. The complete model one would be represented as follows: (4.20) ij k ij ij k ij k k ij ij k ij k ij ij ij k ij ij ij ij ij k ij k ij iju MTH MTH AGE AGE PUR PUR GEN GEN INC INC PEN 12 2 ) ( 1 ) ( ) ( 12 4 2 ) ( 1 ) ( ) ( 8 ) ( 1 ) ( 2 ) ( 8 4 2 ) ( 1 ) ( 2 ) ( 6 ) ( 1 ) ( ) ( ) ( 0 where the variables are explained in Table 4.3 Penetration Model II The Penetration Model II is very similar to the market penetration model I, except that the regions are included in the model as dummy variables. Therefore in this model we will have fewer equations, one for each product form. Let us again define the Market Penetration Model II by PEN, which is also a number between zero and one and using iX as shown in equation (4.19). Now iX includes the additional regional variable added to equation (4.20). Note that the j subscripts for the regions are now dropped:

PAGE 100

80 (4.21) i kk i i k i k i i k i k k i i k i k i i i k i i i i i k i k i iu REG REG MTH MTH AGE AGE PUR PUR GEN GEN INC INC PEN 12 2 9 2 ) ( 1 ) ( ) ( 25 ) ( 1 ) ( ) ( 12 4 2 ) ( 1 ) ( ) ( 8 ) ( 1 ) ( 2 ) ( 8 4 2 ) ( 1 ) ( 2 ) ( 6 ) ( 1 ) ( ) ( ) ( 0 Table 4.3. Variables for the Market Penetration Model I where the variables are explained in table 4.4. Buyer Frequency Models The frequency models follow the same stru cture as the market penetration models, with the two approaches. Therefore the structures are Buyer Frequency Model I with i j equations, and Buyer Frequency model II with i equations. INCOME INC2 = ($25,000 $49,999 = 1) or (otherwise = 0) INC3 = ($50,000 $74,999 = 1) or (otherwise = 0) INC4 = ($75,000 or more = 1) or (otherwise = 0) GENDER GEN2 = (male = 0) and (female = 1) PURPOSE PUR2 = (self = 0) and (gift = 1) AGE AGE2 = (25 – 50 = 1) or (otherwise = 0) AGE3 = (50 – 75 = 1) or (otherwise = 0) AGE4 = (75 or more = 1) or (otherwise = 0) SEASON MTH2 = (February = 1) or (otherwise = 0) MTH3 = (March = 1) or (otherwise = 0) MTH4 = (April = 1) or (otherwise = 0) MTH5 = (May = 1) or (otherwise = 0) MTH6 = (June = 1) or (otherwise = 0) MTH7 = (July = 1) or (otherwise = 0) MTH8 = (August = 1) or (otherwise = 0) MTH9 = (September = 1) or (otherwise = 0) MTH10 = (October = 1) or (otherwise = 0) MTH11 = (November = 1) or (otherwise = 0) MTH12 = (December = 1) or (otherwise = 0)

PAGE 101

81 Table 4.4. Variables for the Market Penetration Model II Frequency Model I As stated earlier, a household must have ma de at least one tran saction per month to be a buyer. For the frequency model, the lower constraint is one, since there must be at least one transaction to be a buyer, and theoretically there is no upper limit. Let iX represent income, gender, purpose, age, and seasonal monthly expenditures; let PRT be the price per transaction and IMR be the inverse mills ratio. Then the general frequency model I would be expressed as follows: INCOME INC2 = ($25,000 $49,999 = 1) or (otherwise = 0) INC3 = ($50,000 $74,999 = 1) or (otherwise = 0) INC4 = ($75,000 or more = 1) or (otherwise = 0) GENDER GEN2 = (male = 0) and (female = 1) PURPOSE PUR2 = (self = 0) and (gift = 1) AGE AGE2 = (25 – 50 = 1) or (otherwise = 0) AGE3 = (50 – 75 = 1) or (otherwise = 0) AGE4 = (75 or more = 1) or (otherwise = 0) SEASONALITY MTH2 = (February = 1) or (otherwise = 0) MTH3 = (March = 1) or (otherwise = 0) MTH4 = (April = 1) or (otherwise = 0) MTH5 = (May = 1) or (otherwise = 0) MTH6 = (June = 1) or (otherwise = 0) MTH7 = (July = 1) or (otherwise = 0) MTH8 = (August = 1) or (otherwise = 0) MTH9 = (September = 1) or (otherwise = 0) MTH10 = (October = 1) or (otherwise = 0) MTH11 = (November = 1) or (otherwise = 0) MTH12 = (December = 1) or (otherwise = 0) REGION REG2 = (Middle Atlantic = 1) or (otherwise = 0) REG3 = (East North Central = 1) or (otherwise = 0) REG4 = (West North Central = 1) or (otherwise = 0) REG5 = (South Atlantic = 1) or (otherwise = 0) REG6 = (East South Central = 1) or (otherwise = 0) REG7 = (West South Central = 1) or (otherwise = 0) REG8 = (Mountain = 1) or (otherwise = 0) REG9 = (Pacific = 1) or (otherwise = 0)

PAGE 102

82 (4.22) ij ij ij iju IMR PRT X FRQ FRQ if 0*ijFRQ 0ijFRQ if 0*ijFRQ, which is the classical Tobit model described on the previous section. The complete model I would be represented as follows: (4.23) ij ij ij k ij ij k ij k k ij ij k ij k ij ij ij k ij ij ij ij ij k ij k ij iju IMR PRT MTH MTH AGE AGE PUR PUR GEN GEN INC INC FRQ ) ( 1 ) ( 1 12 2 ) ( 1 ) ( ) ( 12 4 2 ) ( 1 ) ( ) ( 8 ) ( 1 ) ( 2 ) ( 8 4 2 ) ( 1 ) ( 2 ) ( 6 ) ( 1 ) ( ) ( ) ( 0 where the definition of the variables is the same as presented in Table 4.3, with two additional variables: PRT and IMR, for pri ce per transaction and Inverse Mills Ratio, respectively. The IMR variable is used to ta ke into account the factors that influence a household for becoming a buyer. A Probit model will be run in order to obtain the IMR and use it as a variable in the frequency model. The IMR is calculated by calculating the ratio of the density function to the distribution function. Frequency Model II The definition of the buyer frequency model II is FRQ, for frequency, which is the number of times or transactions a househol d makes in a given period of time. Again, using iX with the regions added, PRT, and IM R, the general Frequency Model II is expressed as follows: (4.24) i i i iu IMR PRT X FRQ FRQ if 0*iFRQ 0iFRQ if 0*iFRQ

PAGE 103

83 which is the classical Tobit model described on the previous section. The complete model II would be represented with the j subscript now dropped as follows: (4.25) u IMR PRT REG REG MTH MTH AGE AGE PUR PUR GEN GEN INC INC FRQi i kk i i k i k i i k i k k i i k i k i i i k i i i i i k i k i i ) ( 1 ) ( 1 12 2 9 2 ) ( 1 ) ( ) ( 25 ) ( 1 ) ( ) ( 12 4 2 ) ( 1 ) ( ) ( 8 ) ( 1 ) ( 2 ) ( 8 4 2 ) ( 1 ) ( 2 ) ( 6 ) ( 1 ) ( ) ( ) ( 0 where the definition of the variables is the same as presented in Table 4.4, with two additional variables: PRT and IMR, for price per transaction and inverse mills ratio respectively.

PAGE 104

84 CHAPTER 5 EMPIRICAL RESULTS The following chapter presents and discusse s the empirical results for both demand models developed in Chapter 4. Each of the demand models has two components, the market penetration and the buyer frequency anal ysis. In order to present the results, an introduction to the data usage is described, including an explanation of the calculations and justifications for all variables for the models. Then, an introduction to the demand model equations is presented. Finally, th e results for both demand models will be presented, each one including the market penetration and the buyer frequency models. Data Usage The cross sectional time series data set was divided into four main groups: households, expenditures, transactions and buyers. Market penetration, frequency of buying and price were then calculated from these data. The first two are the dependent variables for the market penetration mode l and buyer frequency model for both demand models. Market penetration was calculated by dividing the number of buyers by the number of households. Fre quency of buying was obtained by dividing the number of transactions by the number of buyers (transactions per buye r). Price was calculated by dividing total expenditures by number of trans actions (expenditures per transaction). It was not possible to use price as variable on the market pene tration model since the data set contained information only on buyers, ther efore when calculating the average price per transaction for the market penetration (price), it would al so be considering

PAGE 105

85 observations where the expenditure was zer o and therefore the price would not be defined. Each one of the main four groups was s ubdivided into flower types, regions, purpose of the purchase, seasonality, and dem ographics, which included income, age, and gender. There was seven flower types consider ed in the analysis: indoor, cut-flowers, flower arrangements, non-arrangements, flow ering plants, dry/artif icial, and out-door. Figure 5.1 shows the classification of the different flower type groups. Figure 5.1 Flower Type Groups. Regions were defined accordi ng to nine geographical areas: New England, Middle Atlantic, East North Central, West North Ce ntral, South Atlantic, East South Central, West South Central, Mountain and Pacific. Ta ble 5.1 shows the states belonging to each region. The purpose of the purchase was either for self-use or for gifts. Seasonality was analyzed based on monthly data. Demand Model Equations The first demand model consists of a demand analysis of the flower industry where both regional and product form changes are separated and analyzed individually. This means that data were separated by region an d by product form. After separating the data, the analysis was completed for each one of the sub-division of the data set. The resulting ARRANGEMENTS NON-ARRANGEMENTS CUTFLOWERS FLOWERING FOLIAGE PLANTS DRY/ARTIFICIAL INDOOR OUTDOOR

PAGE 106

86 model had one equation for each region and flower type. With nine regions, and the national average results (i); and with seven ma in flower types (j), the first model will have a total of se venty equations (i j) for the market penetration model and seventy equations for the buyer frequency model. The dependent variable for each equation would be market penetration for region i and product form j, and buyer frequency for region i and product form j. Table 5.1 Distribution of the States for Each Region. New England East North Central South Atlantic Mountain MaineOhioMarylandMontana New HampshireIndianaDelawareWyoming VermontIllinoisWashington D.C.Colorado MassachusettsMichiganVirginiaIdaho Rhode IslandWisconsinWest VirginiaNew Mexico ConnecticutNorth CarolinaNevada South CarolinaArizona FloridaUtah GeorgiaMid-Atlantic West North Central East South Central Pacific New YorkMinnesotaKentuckyWashington New JerseyIowaTennesseeOregon PennsylvaniaMissour iAlabamaCalifornia NebraskaMississippi Kansas North Dakota South DakotaWest South Central Arkansas Louisiana Oklahoma Texas

PAGE 107

87 The second demand model consists of a demand analysis with the data concatenated vertically by region. This rearra ngement of the data allows the inclusion of region as an explanatory variable. Therefore, the resulting model will have one equation for each product form or seven equations fo r the market penetration model and seven equations for the buyer frequency model. These two model approaches are feasible due to the large numb er of observations in the data. The objective of ha ving the two analyses is to compare the results from each model and observe if both models yield the sa me conclusions regarding the behavior of their parameters, direction, significance, and agreement with economic theory. It is expected to observe similar re sults in both demand models. In addition to this, the second model also allows the recording of regional differences in the demand for flowers by comparing each region against the average for all regions. All estimates and data were obtained and stored using TSP software. A copy of the TSP programs can be found in Appendix B. Model I Results Since there are one hundred and forty equa tions for this model, including both market penetration and buyer frequency mo dels, only the national results will be presented; that is, the results for the aver age United States for each one of the flower types (total of 7). The remaining 126 equati ons, 63 for the market penetration model and 63 for the buyer frequency model, which includ e the result equations for each one of the nine remaining regions and seven produc t forms, can be found in Appendix A. Market Penetration Model Results I The dependent variable for this model is market penetration for region i and product form j. As stated above there are a to tal of 70 equations for this model. In this

PAGE 108

88 section, it will only be presented the result s aggregated over the regions showing seven equations, one corresponding to market pene tration by product form. Table 5.2 provides the number of observations, the log likelihood value of the Tobit estimates, and the tvalue at the 95 % confidence level. Table 5.2 General Statistical Information A bout the Market Penetr ation Model by Flower Type. The number of observations was the same fo r all flower types since the data set was only divided by flower type. The log likelihood functions are reported for all seven equations. The t-value at a confidence of 95 % is 1.96, which means that if the t-value for the parameters were greater than 96 1, one would reject the null hypothesis that the parameter is equal to zero and hence, conc lude that the parameter is statistically significant at 95 % confidence level. Tables 5.3, 5.4, 5.5, and 5.6 show the Tobit parameter estimates and th eir corresponding t-values. At a 95 % confidence interval, all of the pa rameters are significant for cut-flowers, and all except age between 2539 years old for indoor. All dummy variables are based on the average of all its categorie s; one can easily calculate the value of the parameter for the category not included on the model by simple calculating the negative sum of the rest of the categories for each dummy class. For example, to calculate the first category of No. Observations Log Likelihood Function t-value at 95% Indoor 921639536.11.96 Cut-Flowers 921639488.21.96 Arrangements 921628336.81.96 Non-arragements 921639995.01.96 Plants 921638816.61.96 Artificial 921628987.81.96 Outdoor 921629308.41.96

PAGE 109

89 income, DINC1=-(DINC2+DINC3+DINC4). The intercept then represents the average over all variables in the models. Table 5.3 Market Penetrati on Parameter Estimates and TValues for Indoor and Cut-Flowers. The negative sign means that there exists an inverse relationship between the dummy variable and the dependent variable, while a positive sign implies a direct relationship. For example, DGEN2 (female) parameter estimate is positive and significant. Since these are Tobit estimates one cannot immediately infer that the coefficient indicates a specific change in the units of mark et penetration. One obviously has to deal with the number or probability of not buying. This is completed later in the simulation chapter. Note that the t-value provid es a test of the difference of each estimate BETA T-VALUE BETA T-VALUE C 0.0018982.039800.0007765.28962 DINC2 (25/50) 0.000184.447770.000115.53435 DINC3 (50/75) -0.00079-19.85381-0.00032-16.06191 DINC4 (75+) 0.0007919.801320.0004120.87022 DGEN2 (Female) 0.0012956.059320.0005042.78966 DPUR2 (Gifts) 0.0002812.295260.0005849.80004 DAGE2 (25/39) -0.00003-0.823270.000084.15290 DAGE3 (40/55) 0.0004912.402080.0003819.63520 DAGE4 (55+) 0.0015038.002460.0005327.26822 DMTH2 (February) 0.000445.726370.0005715.06096 DMTH3 (March) 0.000314.071070.000194.96892 DMTH4 (April) 0.000678.835420.000215.44479 DMTH5 (May) 0.0010814.372910.0005013.47163 DMTH6 (June) -0.00032-4.22323-0.00012-3.14181 DMTH7 (July) -0.00056-7.29076-0.00021-5.58443 DMTH8 (August) -0.00054-7.03652-0.00016-4.18827 DMTH9 (September) -0.00047-6.08776-0.00018-4.78738 DMTH10(October) -0.00039-5.08755-0.00013-3.39330 DMTH11(November) -0.00024-3.18629-0.00015-3.93865 DMTH12(December) 0.000547.06612-0.00028-7.27757 SIGMA 0.00216130.149980.00105121.21377 INDOOR CUT-FLOWERS

PAGE 110

90 from the average. One must know the varian ce-covariance matrix in order to test the difference between two levels within a dummy class. Table 5.4 Market Penetrati on Parameter Estimates and TValues for Flower Arrangements and Non-Arrangements. For all flower types, female buyers in crease market penetration. When buyer purpose was for gifts, market penetration in creased for indoor, cut-flowers, arrangements and non-arrangements, while it decreased for pl ants, dry/artificial and outdoor. In general we can observe that market penetration genera lly increases with age and equally that the age effect is statistically significant. Later simulations will show the richness of these conclusions. BETA T-VALUE BETA T-VALUE C -0.00008-11.060840.0005757.59825 DINC2 (25/50) 0.000099.230020.000073.96481 DINC3 (50/75) -0.00013-12.39562-0.00024-14.54237 DINC4 (75+) 0.0001413.068780.0003219.29407 DGEN2 (Female) 0.0002133.398740.0003940.01904 DPUR2 (Gifts) 0.0005170.889610.0003333.97230 DAGE2 (25/39) 0.000098.557190.000052.90277 DAGE3 (40/55) 0.0001716.894260.0003018.43897 DAGE4 (55+) 0.0002221.398160.0004225.65233 DMTH2 (February) 0.0002513.107970.0004213.30631 DMTH3 (March) -0.00003-1.464100.000216.53076 DMTH4 (April) 0.000104.866880.000165.05774 DMTH5 (May) 0.0002613.562620.0003611.37875 DMTH6 (June) -0.00010-4.95229-0.00007-2.06731 DMTH7 (July) -0.00012-5.94895-0.00015-4.58091 DMTH8 (August) -0.00012-5.64920-0.00010-3.21648 DMTH9 (September) -0.00010-5.01335-0.00013-4.06536 DMTH10(October) -0.00006-3.05417-0.00008-2.64504 DMTH11(November) -0.00002-1.08873-0.00013-3.97319 DMTH12(December) 0.000021.24976-0.00030-9.13310 SIGMA 0.0004896.473860.00088119.37811 ARRANGE NON-ARRAN

PAGE 111

91 Table 5.5 Market Penetrati on Parameter Estimates and TValues for Plants and Dry/Artificial Flowers. The seasonal effects differ from one flow er type to another with each monthly parameter being compared to an average over a twelve month cycle. In order to illustrate seasonal effects, Figures 5.2 and 5.3 present seasonal parameters of cut-flowers, plants, dry/artificial, and out-door. For cut-flowers, market penetration was higher than average during the period from February to May, wh ich coincides with the ValentineÂ’s and the MotherÂ’s Day buying occasions. Each of the se asonal patterns points to the considerable downtrend in all forms during the fall months. Th is has been a long term pattern in the flower industry and clearly highlights a mark eting challenge to stimulate the demand during the fall months. BETA T-VALUE BETA T-VALUE C 0.0006144.29153-0.00002-2.34693 DINC2 (25/50) 0.000125.074700.000075.20278 DINC3 (50/75) -0.00043-17.92673-0.00030-21.51743 DINC4 (75+) 0.0003314.139350.000085.80097 DGEN2 (Female) 0.0007454.216710.0005767.79331 DPUR2 (Gifts) -0.00017-12.51262-0.00008-10.69488 DAGE2 (25/39) 0.000010.235720.000010.66271 DAGE3 (40/55) 0.0002711.710720.0001410.80562 DAGE4 (55+) 0.0008938.483530.0005139.70127 DMTH2 (February) -0.00007-1.565380.000062.41253 DMTH3 (March) 0.000102.311030.000114.37852 DMTH4 (April) 0.0005612.669870.000072.78702 DMTH5 (May) 0.0006715.300270.000218.45805 DMTH6 (June) -0.00016-3.48069-0.00014-5.49312 DMTH7 (July) -0.00035-7.67799-0.00016-6.13371 DMTH8 (August) -0.00039-8.61712-0.00012-4.52121 DMTH9 (September) -0.00038-8.33990-0.00005-1.87588 DMTH10(October) -0.00031-6.81072-0.00001-0.29940 DMTH11(November) -0.00017-3.845300.000041.69651 DMTH12(December) 0.0008519.229620.000072.60425 SIGMA 0.00124122.411310.0006399.35858 PLANTS DRY

PAGE 112

92 Table 5.6 Market Penetration Parameter Estimates and T-Values for Outdoor Flowers. For plants one can observe that from March to May and in December the market penetration is above average with purchases When purchases occur during the remainder of the months, market penetra tion is below average. For dry/artificial, market penetration increases when purchases occur from February to May and November-December. In the case of outdoor flowers, market penetration increases above average for purchases made from March to June when considering outdoor gardening takes place. As it can be seen in the graphics, seasonal effects i ndicate that market penetra tion is highly impacted by calendar occasions. The significance of the “s igma” variable means that for the data truncation, one cannot ignore the lower limit level of zero; the estimation method must deal with the asymptotic distributi on of the data (i.e., Tobit model). BETA T-VALUE C 0.000123.60365 DINC2 (25/50) 0.000326.11552 DINC3 (50/75) -0.00058-10.66405 DINC4 (75+) 0.000417.65615 DGEN2 (Female) 0.0011336.45112 DPUR2 (Gifts) -0.00095-30.78801 DAGE2 (25/39) 0.000040.83202 DAGE3 (40/55) 0.0007514.50293 DAGE4 (55+) 0.0016632.18746 DMTH2 (February) -0.00070-6.78128 DMTH3 (March) 0.000323.28148 DMTH4 (April) 0.0018419.06963 DMTH5 (May) 0.0035837.59296 DMTH6 (June) 0.0012712.97691 DMTH7 (July) -0.00017-1.64839 DMTH8 (August) -0.00073-7.04090 DMTH9 (September) -0.00033-3.20333 DMTH10(October) -0.00049-4.71796 DMTH11(November) -0.00135-12.43739 DMTH12(December) -0.00168-15.17570 SIGMA 0.00270117.13395 OUTDOOR

PAGE 113

93 Figure 5.2 Market Penetration Seasona lity for Cut-Flowers and Plants. Buyer Frequency Model Results I For the first demand model, the buyer fre quency data set was arranged in the same manner as the market penetration for a total of 70 equations. Again, aggregated results are reported in the text. The dependent va riable is the buyer frequency across seven product forms. Table 5.7 provides the number of observations, the log likelihood value of the Tobit function, and the table t-value for a 95 % confidence level. These results are truncated at zero and will be adjusted in the simulation se ction by adding back the one that was subtracted for estimation purposes.

PAGE 114

94 Figure 5.3 Market Penetrati on Seasonality for Dry/Artific ial and Outdoor flowers. Tables 5.8, 5.9, 5.10, and 5.11 show the Tobit parameter estimates and their corresponding t-values for the buyer frequenc y models. The only continuous variable is price and, as expected, the negative sign on price is consistent with economic theory across all seven models. Price parameters ar e statistically significant except for flower arrangements, although price in the arra ngements equation has a negative sign. As explained in Chapter 4, to explain the fre quency of buying model one needs to take into consideration the probability that a household is a buyer. In order to incorporate this probability in the model, the Inverse Mill s Ratio was introduced as an explanatory

PAGE 115

95 variable. Mills was obtained from a Probit an alysis of the market penetration model. Mills is the ratio of the density to the di stribution function. Th e main purpose of the MillsÂ’ variable is to take into consideration the probabili ty that a house hold would be a buyer and if that would have an effect on the frequency of buying. A significant parameter for the Mills variable says that there are some factors that influence a household into becoming a buyer and that these factors have an impact on the frequency of buying. Table 5.7 General Statistical Information About the Buyer Frequency Model by Flower Type. Again, when interpreting the variables, the Tobit estimates cannot immediately infer that the coefficient mean s a specific change in unit of buyer frequency. In order to calculate the margins truncation must be consid ered. This part of the analysis will be presented in the simulation chapter. Buyer frequency increases when females make the purchase. On the other hand, one sees that frequency of buying or number of transactions during a given period is reduced if the purpose of the purchase is for gifts. This makes practical sense; it may be inferred that if a household were purchasing flowers for gifts, the frequency would be for those gifts. On the other hand, households that buy flowers for self-use, especially females, would increas e the frequency of buying. That is what the positive sign of the parameter estimates show. No. Observations Log Likelihood Function t-value at 95% Indoor 8472-7176.51.96 Cut-Flowers 7439-6447.91.96 Arrangements 4784-3108.21.96 Non-arragements 7317-6693.11.96 Plants 7531-6825.51.96 Artificial 5150-6536.71.96 Outdoor 6732-8310.01.96

PAGE 116

96 Table 5.8 Buyer Frequency Parameter Estimates and T-values for Indoor and Cut-Flowers. Table 5.9 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements and Non-Arrangements. BETA T-VALUE BETA T-VALUE C 0.4671534.055780.3299419.38316 DINC2 (25/50) -0.02918-2.63885-0.05705-4.49594 DINC3 (50/75) -0.04907-4.38731-0.02130-1.63897 DINC4 (75+) 0.030522.583420.054154.12829 DGEN2 (Female) 0.0836611.965540.018141.86079 DPUR2 (Gifts) -0.08433-10.71380-0.10125-7.75658 DAGE2 (25/39) 0.030042.70426-0.05501-4.22437 DAGE3 (40/55) 0.098778.617490.082136.13268 DAGE4 (55+) 0.064285.650820.1807713.84435 DMTH2 (February) 0.048652.337120.010400.43518 DMTH3 (March) 0.036811.77974-0.00391-0.16248 DMTH4 (April) -0.00610-0.294420.005120.21386 DMTH5 (May) 0.066983.265250.029701.25381 DMTH6 (June) 0.037591.801670.054392.24714 DMTH7 (July) -0.02509-1.173170.004140.16698 DMTH8 (August) 0.017280.814250.018800.76335 DMTH9 (September) -0.02379-1.12269-0.02273-0.91813 DMTH10(October) -0.02439-1.16030-0.04930-2.01491 DMTH11(November) -0.05841-2.76032-0.04620-1.87301 DMTH12(December) -0.04367-2.082350.016680.65949 PRT -0.00749-8.97614-0.00385-4.53025 MILLS -0.14437-2.63736-0.32775-6.76708 SIGMA 0.56473112.812750.6027097.67298 INDOOR CUT-FLOWERS BETA T-VALUE BETA T-VALUE C -0.89681-10.990370.3240215.36399 DINC2 (25/50) 0.085903.86210-0.05591-3.91762 DINC3 (50/75) -0.06582-2.92324-0.02759-1.88529 DINC4 (75+) 0.091534.285860.064164.34872 DGEN2 (Female) 0.2125011.398640.001850.16315 DPUR2 (Gifts) 0.432408.28950-0.13220-9.87354 DAGE2 (25/39) 0.038361.57376-0.07232-4.89176 DAGE3 (40/55) 0.176526.516510.100306.53443 DAGE4 (55+) 0.193727.113450.2050313.73049 DMTH2 (February) 0.152384.001220.022100.82702 DMTH3 (March) -0.00926-0.226810.003570.13221 DMTH4 (April) 0.019630.489760.011780.44018 DMTH5 (May) 0.215045.638310.042381.59921 DMTH6 (June) -0.06510-1.517790.062002.28621 DMTH7 (July) -0.06258-1.44656-0.00675-0.24184 DMTH8 (August) 0.009270.217030.007570.27401 DMTH9 (September) -0.11700-2.64761-0.01178-0.42123 DMTH10(October) -0.07125-1.69463-0.05870-2.13839 DMTH11(November) -0.07966-1.90386-0.05228-1.88237 DMTH12(December) 0.122673.147210.004050.14030 PRT -0.00059-1.00038-0.00681-4.92238 MILLS 0.471914.92075-0.35608-6.42195 SIGMA 0.6841350.100730.6634793.10333 ARRANGE NON-ARRAN

PAGE 117

97 Table 5.10 Buyer Frequenc y Parameter Estimates and T-values for Plants and Dry/Artificial. Table 5.11 Buyer Frequency Parameter Estimates and T-values for Outdoor. BETA T-VALUE BETA T-VALUE C 0.148435.557540.261632.14261 DINC2 (25/50) 0.014720.885800.092682.64345 DINC3 (50/75) -0.05689-3.47096-0.12407-3.34786 DINC4 (75+) 0.024691.38077-0.07991-1.85759 DGEN2 (Female) 0.104237.777900.473556.32109 DPUR2 (Gifts) -0.18629-19.72382-0.07949-4.31817 DAGE2 (25/39) 0.055433.272930.179164.98435 DAGE3 (40/55) 0.129896.866450.177694.11250 DAGE4 (55+) 0.068453.426090.038860.85758 DMTH2 (February) 0.009150.311330.118261.99947 DMTH3 (March) 0.077342.676940.147782.49747 DMTH4 (April) 0.110253.836020.081131.36975 DMTH5 (May) 0.154405.286110.188653.22120 DMTH6 (June) -0.02159-0.72075-0.10731-1.67409 DMTH7 (July) -0.08268-2.60253-0.11748-1.84007 DMTH8 (August) -0.07693-2.434810.121071.94987 DMTH9 (September) -0.05561-1.76568-0.08193-1.33111 DMTH10(October) -0.03724-1.21538-0.06745-1.12265 DMTH11(November) -0.10773-3.50275-0.09771-1.61730 DMTH12(December) 0.074302.59229-0.10565-1.76076 PRT -0.00986-9.25510-0.02767-15.43293 MILLS -0.01571-0.201510.107370.66957 SIGMA 0.7112389.110631.2078376.83770 PLANTS DRY BETA T-VALUE C 0.4972212.07907 DINC2 (25/50) 0.011530.48023 DINC3 (50/75) -0.02108-0.89247 DINC4 (75+) 0.055602.22310 DGEN2 (Female) 0.111666.55921 DPUR2 (Gifts) -0.37451-24.59005 DAGE2 (25/39) 0.138455.84487 DAGE3 (40/55) 0.135104.79686 DAGE4 (55+) 0.087102.95014 DMTH2 (February) -0.15762-3.46481 DMTH3 (March) 0.219425.17302 DMTH4 (April) 0.388868.98733 DMTH5 (May) 0.6058313.20435 DMTH6 (June) 0.363718.59079 DMTH7 (July) -0.01621-0.37497 DMTH8 (August) -0.11594-2.52297 DMTH9 (September) -0.06010-1.37049 DMTH10(October) -0.07306-1.62751 DMTH11(November) -0.24872-4.72117 DMTH12(December) -0.61470-10.07254 PRT -0.01419-10.34917 MILLS 0.025140.28356 SIGMA 1.0226992.83852 OUTDOOR

PAGE 118

98 The seasonal effects differ from one flower type to another compared to an average over the twelve months cycle. In order to illu strate seasonal effects on frequency, Figures 5.4and 5.5 present seasonal parameters of cutflowers, plants, dry/artificial, and outdoor. In the case of cut-flowers, buyer frequency is above average in the months of February, April to August and December. Frequency be ing above average during these months is highly related to special calenda r occasion (ValentineÂ’s, MotherÂ’s Day, Christmas, etc.). In the case of plants, a similar tendency of buyer frequency increasing for the period of March through May and December correspond to specific calendar occasions. Figure 5.4 Buyer Frequency Seasonality for Cut-Flowers and Plants.

PAGE 119

99 Buyer frequency for dry/artificial flowers in creases in the period from February to May and in the month of August. Outdoor frequency of purchasin g increases if the purchase was made in the period from Marc h through June. As shown in the Figures, seasonal variations in flowers, frequency of buying for different flower types is highly related by calendar, that is, occasions that occur during the same month each year. Figure 5.5 Buyer Frequency Seasonality fo r Dry/Artificial and Outdoor Flowers. Model II Results The second type analyses incorporate the nine regions as an independent dummy variable within each model. Hence, the model will have seven equations, one for each

PAGE 120

100 product form. For regional differences, each re gion will be compared to an average over the nine regions. A positive regional parameter would mean that the regionÂ’s market penetration or buying frequency is higher th an the average of the whole country. Market Penetration Model Results II The dependent variable for this model is market penetration for product form j. As stated above, there are a total of 7 e quations, one for each product form. Table 5.12 provides the number of obser vations, the log likelihood valu e of the Tobit function and the t-value at a 95 % confidence level. Table 5.12 General Statistica l Information About the Mark et Penetration Model II by Flower Type. The number of observations 82,944 was the sa me for all flower types as shown in Table 5.12 and the log likelihood functions ar e reported for all seven equations. Tables 5.13, 5.14, 5.15, and 5.16 show the Tobit parameter estimates an d its corresponding tvalues. Most of the parameter estimates for the market penetration model II are significant at the 95% c onfidence level. In the case of i ndoor flowers and cut-flowers, all of the parameters are significant at the 95% level. No. Observations Log Likelihood Function t-value at 95% Indoor 82,944178,525.01.96 Cut-Flowers 82,944133,107.01.96 Arrangements 82,94452,230.11.96 Non-arragements 82,944113,492.01.96 Plants 82,944115,730.01.96 Dry/Artificial 82,94461,811.71.96 Outdoor 82,94485,170.41.96

PAGE 121

101 Table 5.13 Market Penetration Parameter Estimates and T-Values for Indoor and Cut-Flowers. BETA T-VALUE BETA T-VALUE C 0.0003017.25350-0.00103-64.23079 DINC2 (25/50) 0.0006743.637320.0012395.24360 DINC3 (50/75) 0.00205130.341790.0010179.70666 DINC4 (75+) 0.0005119.597790.0003818.50793 DGEN2 (Female) -0.00123-45.36211-0.00059-27.29302 DPUR2 (Gifts) 0.0007428.166260.0005124.61681 DAGE2 (25/39) 0.0002610.078740.0002913.75682 DAGE3 (40/55) 0.0010741.404240.0008943.70466 DAGE4 (55+) 0.0021885.185020.0010752.47922 DMTH2 (February) 0.0008116.433570.0010427.35504 DMTH3 (March) 0.000479.485900.0003910.01941 DMTH4 (April) 0.0010320.968340.0003910.05105 DMTH5 (May) 0.0016333.685420.0008822.89019 DMTH6 (June) -0.00044-8.60491-0.00025-6.21934 DMTH7 (July) -0.00081-15.69637-0.00034-8.34254 DMTH8 (August) -0.00086-16.53993-0.00032-7.76026 DMTH9 (September) -0.00076-14.76946-0.00038-9.15254 DMTH10(October) -0.00055-10.83029-0.00019-4.69202 DMTH11(November) -0.00040-7.77517-0.00029-7.16966 DMTH12(December) 0.0006212.49048-0.00057-13.63343 DREG2(Middle Atlantic) 0.0007317.616100.0009027.77186 DREG3(ENC) 0.0005011.923770.0005015.38483 DREG4(WNC) -0.00049-11.11033-0.00053-14.87238 DREG5(S. Atlantic) 0.0004711.146900.0003611.02142 DREG6(ESC) -0.00093-20.70836-0.00131-33.99515 DREG7(WSC) -0.00031-7.08314-0.00036-10.36205 DREG8(Mountain) -0.00075-16.83839-0.00063-17.16324 DREG9(Pacific) 0.0005412.826630.0005316.00743 SIGMA 0.00389294.115740.00286242.29771 INDOOR CUT-FLOWERS

PAGE 122

102 Table 5.14 Market Penetration Parameter Estimates and T-Values for Flower Arrangements and Non-Arrangements. BETA T-VALUE BETA T-VALUE C -0.00311-103.97395-0.00137-82.27114 DINC2 (25/50) 0.0017290.321310.0008064.38560 DINC3 (50/75) 0.0006144.293790.0008870.81152 DINC4 (75+) 0.0003817.575170.0003316.64155 DGEN2 (Female) -0.00036-15.41930-0.00051-24.33036 DPUR2 (Gifts) 0.0003314.971910.0004321.15294 DAGE2 (25/39) 0.0003314.692870.000209.99798 DAGE3 (40/55) 0.0006931.050270.0008039.96622 DAGE4 (55+) 0.0007031.817410.0009547.59449 DMTH2 (February) 0.0007920.104390.0009325.00061 DMTH3 (March) -0.00007-1.671270.0004612.02066 DMTH4 (April) 0.000266.146020.000348.97389 DMTH5 (May) 0.0006917.382620.0007620.35079 DMTH6 (June) -0.00036-7.90986-0.00015-3.79519 DMTH7 (July) -0.00036-7.84560-0.00027-6.63269 DMTH8 (August) -0.00034-7.49080-0.00024-6.00338 DMTH9 (September) -0.00031-6.94757-0.00031-7.76471 DMTH10(October) -0.00014-3.24563-0.00014-3.49529 DMTH11(November) -0.00005-1.08585-0.00030-7.44178 DMTH12(December) 0.000092.03701-0.00071-16.89941 DREG2(Middle Atlantic) 0.0004713.617260.0009730.91661 DREG3(ENC) 0.0004312.603800.0005718.04021 DREG4(WNC) -0.00023-5.98588-0.00055-15.49919 DREG5(S. Atlantic) 0.0004513.168630.0003611.06099 DREG6(ESC) -0.00046-11.47337-0.00150-37.84272 DREG7(WSC) -0.00007-1.82163-0.00040-11.46940 DREG8(Mountain) -0.00030-7.59973-0.00072-19.74160 DREG9(Pacific) 0.000174.808840.0006419.97225 SIGMA 0.00231148.743520.00273222.05986 ARRANGE NON-ARRAN

PAGE 123

103 Table 5.15 Market Penetration Parameter Estimates and T-Values for Plants and Dry/Arti ficial Flowers. BETA T-VALUE BETA T-VALUE C -0.00154-85.28418-0.00291-107.73935 DINC2 (25/50) -0.00030-23.19363-0.00026-19.93158 DINC3 (50/75) 0.00163116.231990.00178100.93696 DINC4 (75+) 0.0004219.401370.0003215.04559 DGEN2 (Female) -0.00084-35.97500-0.00080-33.14199 DPUR2 (Gifts) 0.0004218.839920.000146.31582 DAGE2 (25/39) 0.000198.502700.000094.02680 DAGE3 (40/55) 0.0008137.135560.0006027.27760 DAGE4 (55+) 0.0016978.471050.0013963.97581 DMTH2 (February) 0.000020.506930.000122.94446 DMTH3 (March) 0.000266.123390.000317.57742 DMTH4 (April) 0.0011528.900120.000225.31123 DMTH5 (May) 0.0013333.837340.0005413.33882 DMTH6 (June) -0.00026-6.04860-0.00032-7.33064 DMTH7 (July) -0.00072-16.11897-0.00040-8.90367 DMTH8 (August) -0.00088-19.35179-0.00034-7.58235 DMTH9 (September) -0.00075-16.65546-0.00013-3.00111 DMTH10(October) -0.00065-14.62924-0.00002-0.56625 DMTH11(November) -0.00031-7.082340.000112.65099 DMTH12(December) 0.0014737.212430.000112.74138 DREG2(Middle Atlantic) 0.0003911.030120.000133.58508 DREG3(ENC) 0.0004613.132200.0005516.35153 DREG4(WNC) -0.00048-12.736790.000164.29376 DREG5(S. Atlantic) 0.0004914.234680.0006419.22109 DREG6(ESC) -0.00070-18.054900.000339.22037 DREG7(WSC) -0.00012-3.211940.000267.53945 DREG8(Mountain) -0.00057-14.74975-0.00071-17.57274 DREG9(Pacific) 0.0007020.34068-0.00023-6.35013 SIGMA 0.00292229.164110.00238162.47096 PLANTS DRY

PAGE 124

104 Table 5.16 Market Penetration Parameter Estimates and TValues for Outdoor Flowers. BETA T-VALUE C -0.00475-119.98363 DINC2 (25/50) -0.00209-80.34914 DINC3 (50/75) 0.0024993.35480 DINC4 (75+) 0.0007517.88952 DGEN2 (Female) -0.00108-24.14127 DPUR2 (Gifts) 0.0007417.16205 DAGE2 (25/39) 0.000296.47271 DAGE3 (40/55) 0.0017641.03134 DAGE4 (55+) 0.0033679.48961 DMTH2 (February) -0.00160-17.82772 DMTH3 (March) 0.0008710.96505 DMTH4 (April) 0.0041056.86159 DMTH5 (May) 0.0068998.76059 DMTH6 (June) 0.0029940.21782 DMTH7 (July) -0.00003-0.35413 DMTH8 (August) -0.00131-14.83185 DMTH9 (September) -0.00019-2.30976 DMTH10(October) -0.00057-6.68502 DMTH11(November) -0.00305-30.59543 DMTH12(December) -0.00425-38.69074 DREG2(Middle Atlantic) 0.000537.76639 DREG3(ENC) 0.000547.88223 DREG4(WNC) -0.00082-10.90657 DREG5(S. Atlantic) 0.0014722.44295 DREG6(ESC) -0.00104-13.65922 DREG7(WSC) -0.00016-2.19071 DREG8(Mountain) -0.00150-19.37633 DREG9(Pacific) 0.0014922.73308 SIGMA 0.00519222.00956 OUTDOOR

PAGE 125

105 Market penetration increases with female buyers for all flower types; it also increases with gift purchases for indoor pl ants, cut-flowers, flower arrangements and non-arrangements. Market penetration decr eases with gift purchases for plants, dry/artificial, a nd outdoor plants. Figure 5.6 Market Penetration Seasona lity for Cut-Flowers and Plants. This may be attributed to people buying th ese types of flower products for selfconsumption. The seasonal variables and re gional variables can be compared to the overall average of the nine regions and twelve months. A positive parameters implies that market penetration is higher than average wh en purchases are made during that month or

PAGE 126

106 in that region. Figures 5.6 and 5.7 show seas onal variations for the main flower types, and Figures 5.8 and 5.9 show regional variations from the regional mean. Figure 5.7 Market Penetrati on Seasonality for Dry/Artific ial and Outdoor flowers. As shown in Figures 5.6 and 5.7, seasonal variations of the pa rameters are highly related to special calendar occasions. Cu t flowers have higher than average market penetration from the period of February thr ough May. Plant market penetration is higher than average from March through May and De cember; Dry/artificial market penetration is higher than average from February th rough May and in August; and Outdoor plants have a higher market penetration than av erage from March through June. These results

PAGE 127

107 are very similar to the results obtained in the first model developed. The common characteristic of both is that the nature of the seasonality depends on calendar occasions. Regional changes are shown in Figures 5.8 and 5.9; the graphs represent deviations from the mean of all regions. Figure 5.8 Market Penetration Regional Changes for Cut-Flowers and Plants.

PAGE 128

108 The regions with a positive parameter have a market pe netration higher than the overall nine-region average. Th e highest increase in market penetration o ccurs in New England, for all product forms. Most of the re gional parameters are significant at a 95 % confidence interval, meaning that they are st atistically different from the mean. If the parameters are positive, then they are statis tically significantly higher than the mean; and if the parameters are negative, then they are statistically significantly lower than the regional mean (i.e., if their tvalues are greater than 1.96, at 95% confidence interval). Figure 5.9 Market Penetration Regional Change s for Dry/Artificial and Outdoor Flowers.

PAGE 129

109 Buyer Frequency Model Results II For the second demand model, the buyer frequency data set was arranged in the same manner as the market penetration. Th e dependent variable on this model is the buyer frequency of purchasing by flower t ype. A total of seven equations will be presented, each one corresponding to the buye r frequency of purchasing by product form. Regional changes are included in the model as an indepe ndent dummy variable. Each region is compared to an overall average ov er the nine regions; th erefore the parameter estimates are represented as deviations from the regional mean. Table 5.17 provides the number of observations, the l og likelihood value of the Tobit function, and the t-value at the 95 % confidence level. Table 5.17 General Statistical Information About the Buyer Frequency Model II by Flower Type. The log likelihood functions are reported fo r all seven equations. Tables 5.18, 5.19, 5.20, and 5.21 show the Tobit parameter estim ates and its corresponding t-values for the buyer frequency model. Price has the same in terpretation presented for Model I, as does the Mills Ratio and sigma. The variables XGENPRT and XPURPRT are the interaction between females and price and gift purchases and price. The seasonal variations of the variables are recorded in Figures 5.10 and 5.11. No. Observations Log Likelihood Function t-value at 95% Indoor 47,698-54,055.81.96 Cut-Flowers 34,715-33,550.31.96 Arrangements 14,522-8,058.91.96 Non-arragements 30,222-29,334.61.96 Plants 30,785-29,282.41.96 Dry/Artificial 16,855-22,472.71.96 Outdoor 26,247-37,125.81.96

PAGE 130

110 Table 5.18 Buyer Frequency Parameter Estimates and T-values for Indoor and Cut-Flowers. BETA T-VALUE BETA T-VALUE C -0.16829-3.97844-0.70409-6.86648 DINC2 (25/50) -0.15460-12.72413-0.09612-3.10974 DINC3 (50/75) 0.2830015.648300.136825.16856 DINC4 (75+) 0.035762.774020.005770.29705 DGEN2 (Female) -0.10465-8.40950-0.05056-2.93337 DPUR2 (Gifts) 0.062385.048150.1680611.64094 DAGE2 (25/39) 0.032832.62211-0.06371-3.58361 DAGE3 (40/55) 0.2295713.840400.246708.79855 DAGE4 (55+) 0.2199012.966060.3596413.08805 DMTH2 (February) 0.097454.821980.129154.34110 DMTH3 (March) 0.098134.905610.054832.01803 DMTH4 (April) 0.105615.228520.063902.32107 DMTH5 (May) 0.2455111.943300.210827.40622 DMTH6 (June) -0.04622-2.188020.007650.26738 DMTH7 (July) -0.09141-4.16479-0.05210-1.79230 DMTH8 (August) -0.03923-1.773630.005760.19811 DMTH9 (September) -0.06803-3.10780-0.04727-1.60061 DMTH10(October) -0.08867-4.17736-0.09717-3.42785 DMTH11(November) -0.09273-4.35913-0.12268-4.17129 DMTH12(December) -0.00802-0.39420-0.02795-0.87221 DREG2(Middle Atlantic) 0.132376.821310.255377.38630 DREG3(ENC) 0.119796.313790.158034.91927 DREG4(WNC) -0.19628-9.51825-0.27559-8.65993 DREG5(S. Atlantic) 0.151697.980990.206076.75329 DREG6(ESC) -0.02861-1.19245-0.32061-6.20675 DREG7(WSC) 0.042912.422650.051792.06190 DREG8(Mountain) -0.22900-9.73888-0.28979-7.03051 DREG9(Pacific) 0.119006.442640.186076.12228 PRT -0.01303-20.49436-0.01401-11.75231 MILLSPROB 0.169002.875480.279532.82170 XGENPRT -0.00112-2.280920.002243.97435 XPURPRT 0.004186.707980.006915.94155 SIGMA 1.20901204.394281.30676143.47365 INDOOR CUT-FLOWERS

PAGE 131

111 Table 5.19 Buyer Frequency Parameter Estimates and T-values for Flower Arrangements and Non-Arrangements. BETA T-VALUE BETA T-VALUE C -6.52128-7.69676-0.70239-3.72391 DINC2 (25/50) 1.567525.64227-0.19879-5.06178 DINC3 (50/75) 0.738417.523470.065921.62943 DINC4 (75+) 0.403135.16323-0.03827-1.36272 DGEN2 (Female) -0.27791-5.16433-0.03319-1.40992 DPUR2 (Gifts) 0.358828.647090.1757910.22874 DAGE2 (25/39) 0.254063.85151-0.10853-4.89743 DAGE3 (40/55) 0.775315.978580.237925.49343 DAGE4 (55+) 0.794606.664010.388539.12875 DMTH2 (February) 0.789996.610350.049721.16600 DMTH3 (March) -0.03593-0.548210.058631.72405 DMTH4 (April) 0.240443.439500.083192.42497 DMTH5 (May) 0.819198.070310.163904.12832 DMTH6 (June) -0.35452-4.039810.037301.10187 DMTH7 (July) -0.38026-4.44695-0.00654-0.18626 DMTH8 (August) -0.32519-3.867540.024710.70823 DMTH9 (September) -0.30255-3.76156-0.02086-0.57534 DMTH10(October) -0.28221-4.00140-0.08201-2.42058 DMTH11(November) -0.21895-3.18475-0.09413-2.56727 DMTH12(December) 0.398036.63332-0.11021-2.34723 DREG2(Middle Atlantic) 0.798106.751270.206173.48509 DREG3(ENC) 0.710085.589100.138512.70035 DREG4(WNC) -0.30354-3.90413-0.35333-7.81423 DREG5(S. Atlantic) 0.690935.637060.198734.40045 DREG6(ESC) -0.74554-5.46911-0.23692-2.60642 DREG7(WSC) 0.079601.433150.065862.04980 DREG8(Mountain) -0.70727-5.79374-0.24308-3.84058 DREG9(Pacific) 0.316283.912280.173203.49160 PRT -0.00861-3.75966-0.01889-10.87195 MILLSPROB 2.574495.618920.148680.90189 XGENPRT 0.000070.092000.005234.61751 XPURPRT 0.009124.055000.006493.85719 SIGMA 1.5638556.356271.44436129.66988 ARRANGE NON-ARRAN

PAGE 132

112 Table 5.20 Buyer Frequency Parameter Estimates and T-values for Plants and Dry/Artificial. BETA T-VALUE BETA T-VALUE C -1.39956-8.86674-1.83478-2.09130 DINC2 (25/50) -0.37629-24.95923-0.07668-1.73336 DINC3 (50/75) 0.380097.483541.051643.45993 DINC4 (75+) 0.107324.038530.139521.66350 DGEN2 (Female) -0.17929-7.43191-0.34498-3.03814 DPUR2 (Gifts) 0.109415.695290.110992.52433 DAGE2 (25/39) 0.094664.251560.202324.09193 DAGE3 (40/55) 0.291707.419940.274202.09024 DAGE4 (55+) 0.321396.658830.356421.73978 DMTH2 (February) -0.01941-0.586250.061650.91447 DMTH3 (March) 0.150514.663760.305184.11997 DMTH4 (April) 0.354868.861290.231743.34431 DMTH5 (May) 0.4505910.580590.502626.20197 DMTH6 (June) -0.05787-1.68545-0.32434-3.89796 DMTH7 (July) -0.19378-4.77897-0.16419-1.90284 DMTH8 (August) -0.23357-5.376510.014890.17509 DMTH9 (September) -0.23271-5.78668-0.08030-1.14748 DMTH10(October) -0.16185-4.22737-0.04835-0.72436 DMTH11(November) -0.19533-5.51855-0.08562-1.25997 DMTH12(December) 0.356418.91485-0.24972-3.63334 DREG2(Middle Atlantic) 0.240446.841660.093251.07902 DREG3(ENC) 0.286567.383000.290152.07555 DREG4(WNC) -0.31953-8.23952-0.18312-2.94338 DREG5(S. Atlantic) 0.256456.720170.304351.94350 DREG6(ESC) -0.25367-5.134650.243313.90482 DREG7(WSC) 0.034541.216280.156952.03973 DREG8(Mountain) -0.30806-6.79904-0.42704-2.64105 DREG9(Pacific) 0.319578.343440.156862.68119 PRT -0.01516-15.01338-0.04363-17.15082 MILLSPROB 0.638114.695600.758401.50893 XGENPRT -0.00120-1.24803-0.00650-2.58752 XPURPRT 0.001691.72459-0.00795-4.40431 SIGMA 1.42112131.987612.27273112.09783 PLANTS DRY

PAGE 133

113 Table 5.21 Buyer Frequency Parameter Estimates and T-values for Outdoor. BETA T-VALUE C -0.71573-5.33577 DINC2 (25/50) -0.67616-24.54811 DINC3 (50/75) 0.3856612.16691 DINC4 (75+) 0.044991.87623 DGEN2 (Female) -0.04277-1.88715 DPUR2 (Gifts) 0.105794.88366 DAGE2 (25/39) 0.203808.42547 DAGE3 (40/55) 0.300768.69668 DAGE4 (55+) 0.248345.82585 DMTH2 (February) -0.28368-5.48445 DMTH3 (March) 0.288757.08566 DMTH4 (April) 0.8531714.84563 DMTH5 (May) 1.2450517.12464 DMTH6 (June) 0.7374415.43962 DMTH7 (July) 0.054461.33954 DMTH8 (August) -0.21305-4.55156 DMTH9 (September) -0.13859-3.33292 DMTH10(October) -0.12228-2.86071 DMTH11(November) -0.62040-9.16956 DMTH12(December) -1.12814-12.46443 DREG2(Middle Atlantic) -0.06614-1.91711 DREG3(ENC) 0.052101.53374 DREG4(WNC) -0.01328-0.31566 DREG5(S. Atlantic) 0.214285.24329 DREG6(ESC) -0.20240-4.51310 DREG7(WSC) 0.003720.10952 DREG8(Mountain) -0.17055-3.71596 DREG9(Pacific) 0.292327.30504 PRT -0.02373-19.42723 MILLSPROB 0.440044.31490 XGENPRT -0.00806-7.11094 XPURPRT -0.00123-1.01324 SIGMA 1.74822161.37775 OUTDOOR

PAGE 134

114 Figure 5.10 Buyer Frequency Seasonal ity for Cut-Flowers and Plants. Most of the seasonal parameters are signi ficant within the 95% confidence interval. This means that with a 95 % confidence level or more, it can be said that the seasonal parameters are statistically different from th e monthly mean (t values greater than 1.96). The results are very similar, and show the high relation be tween seasonality and special calendar occasions. Again, depending in the fl ower type, statistically higher than average buyer frequency values during specia l calendar occasions are observed.

PAGE 135

115 Figure 5.11 Buyer Frequency Seasonality fo r Dry/Artificial and Outdoor Flowers. Regional changes are also obtained from the model. The regional parameter estimates are deviations from the overall regional mean. A positive significant value would say that the region is st atistically significantly higher than its mean. (Significant if t value is greater than 96 1). A negative parameter means th at the regions are lower than its mean; and if its t value is greater than 1. 96 in absolute value then it is statistically significantly lower than its mean. Figures 5.12 and 5.13 show regional changes for the main flower types.

PAGE 136

116 Figure 5.12 Buyer Frequency Regional Changes for Cut-Flowers and Plants. The region with the largest statistical difference from the mean region varies depending on the flower type as shown in the graphs. These results yield similar conclusions, as does the market penetration model. Regional differences may occur due to advertisement programs that create similar consumer perceptions for flower products, and also by intrinsic differences of the de mographic population of a particular region. There are a few regional parameters that are not statistically signifi cantly different from the regional mean, but in general the major ity of the parameters are significant.

PAGE 137

117 Figure 5.13 Buyer Frequency Regi onal Changes for Dry/Artifi cial and Outdoor Flowers. Throughout this chapter the results of the parameter estimates for both demand approaches for market penetration and buy er frequency have been presented and discussed. Both approaches yielded simila r conclusions regarding the significance, direction, and seasonal changes in flowers. Even though the signifi cance of the parameter changed depending on the flower type, seasonal and regional results were very similar. Price was significant in almost all equations and had the expected theoretical negative sign. After comparing seasonal variations in monthly purchases, it was observed that

PAGE 138

118 market penetration and frequency of buying increased during the same monthly periods corresponding to special calenda r occasions (i.e., MotherÂ’s Da y, ValentineÂ’s, etc.). As stated before, since the results are Tobit es timates one cannot immediately infer that the coefficient means a specific change in the units of market pene tration or buying frequency. The number or probability of not buying must be considered. This is completed in the next chapter, simulation analysis.

PAGE 139

119 CHAPTER 6 SIMULATION ANALYSIS The previous chapter presented the parame ter estimates and co rresponding statistics for both methodology models separated into market penetration and buyer frequency. After obtaining these parameter estimates, the results will be used to conduct a simulation analysis on market penetration and buye r frequency; the process of making the simulations is developed in more detail in a following section. The second general model, which incorporated regional changes as independent dummy variables, will be used to perform simulations for both market pene tration and buyer frequency models. The chapter will be divided into five sections. The first section will be an introduction to the simulation analysis showing the impact from changes in the independent variables and how they affect the market penetration and buyer frequency of purchasing flowers. Then, the total number of transactions is expressed in terms of the portion attributed to frequency changes versus the changes in th e number of buyers (m arket penetration). Sections two to five will addr ess these two issues for cut-fl owers, plants, dry/artificial and outdoor plants, respectively. The results from the simulation analysis will be used to suggest specific recommendations to the flower industry. Introduction The main objective of this chapter is to measure responses of the dependent variable to specific changes on one or more independent variables relative to the average household set condition. The first step in the simulation analysis will be to calculate the market penetration and buyer frequency valu es for the average household consumer. In

PAGE 140

120 order to do so, the expected value of market penetratio n and buyer frequency will be calculated when all variables are at the average level (i .e., all binary variables set to zero). Each simulation includes the expected value of the dependent variable when it is at its lower limit and the expected value of the depe ndent variable for positive values as shown in equations 4.114.15 in the methodology ch apter. From this first step a market penetration value and a buyer frequency va lue for the average household will be obtained. These values will be compared with changes in the dependent variable when one independent variable changes and the rest remain at their average value. Market penetration and buyer frequency values will be calculated using the same expectation framework for each independent variable in the model, with changes in one variable at a time. With this method, one can easily see the effects of a particular variable without any confounding effects. When calculating the market penetration va lues for each of the variables, we divided buyers by the number of households. However, the nature of the data was structured in such a way that for each one of the categories of the variables it was divided by the total number of household for all the categories and not just the households in a specific category. Therefore the market penetr ation values, even t hough the total variable reflects the correct number, are lower for each category. There is no way to discern if the value of the market penetration is higher compared to another category because it is higher, or simply because it has a large propor tion of the observations for that variable. For example, for the age category, the market penetration for all four age categories will be accurate; but when calculating the market pe netration for households with less than 25 years of age, the number of buyers was divide d by the total number of households on all

PAGE 141

121 age groups and not those households under 25 years of age. One would not be able to report whether the value is lower or higher co mpared to the rest, only due to changes in market penetration because the first age gr oup changed, or because the proportion of age group one is lower or higher than the rest. Again, this procedure was adopted simply because the nature of how the data were generated. After obtaining the values for market penetration and buyer frequency for each variable, a comparison will be made to the average consumer to record their behavior if only that variable changes. Next, an attemp t will be made to explain what proportion of the total number of transactions is due to th e frequency of buying ve rsus the increment in the number of buyers (market penetration). In order to do so, first the market penetration value will be used and multiply it by the tota l number of households to obtain the total number of buyers (BUY). Once the total nu mber of buyers and the frequency of transactions (FRQ) are obtained, one can calculate the total number of transactions for the average household and for changes a ttributed to a specific variable. (6.1) HWD PEN BUY* The number of buyers will be the penetration value times the total number of households, and the total number of transactions w ill be buyers times the frequency of buying. (6.2) FRQ BUY TRN* If one calculates the total number of trans actions for the average consumer (constant buyers and frequency from average household) the number of transactions for the variable category under discussi on (buyers times frequency for a variable, such as age of 25 years or less) and the number of transacti ons if buyers did not change (buyers constant from average household and frequency from the variable), then the proportion of the

PAGE 142

122 transactions for that variable category resulting from to fre quency of buying can be obtained, as shown in equation 6.3. (6.3) FRQ BUY FRQ BUY FRQ BUY FRQ BUY FRQ TRN 1 1 1) ( %, where BUY is the number of buyers, FRQ is the frequency, and the over ban means the value from the average household, and the s ubscript 1 means the value from the variable category. This approach gives the proportion of th e transactions due to frequency for the variable category. For example, 0.15 for income of $25,000 or less would mean that 15 percent of the total number of transactions made by house hold within the income group of $25,000 or less is due to buyer frequency, and 85 percent is due to market penetration. Also meaningful would be to show what porti on of the total number of transactions is due to frequency of buying for the whole income e ffect, instead of only one of its categories. In order to obtain the whole effect for each va riable (i.e., income, age, purpose, gender, etc), the total number of transactions w ill be calculated for each category of each variable. Hence, resulting in four transaction numbers fo r income, one for each of its categories. Next, the category with the lowest and highest number of transactions will be showed. By selecting the lowest and highest number of transact ions for the income categories, the total income effect would be captured. The same approach can be used to calculate the proportion of the transactions due to frequency of buying from the highest income category to the lowest; hence captu ring the whole income effect. The same methodology can be used for the rest of the va riables. Equation 6.3 can be transformed to fit this approach as follows:

PAGE 143

123 (6.4) m m M M m m m M mBUY FRQ BUY FRQ BUY FRQ BUY FRQ FRQ TRN ( ) ( ( ) ( ) ( %1 2 where FRQ represents frequenc y, BUY number of buyers, a nd the subscripts M and m, represent maximum and minimum number of tran sactions for the variable categories. Simulations For Cut-Flowers The average household had a market pe netration value of 0.00070 and a buyer frequency value of 1.25818. The calculated nu mber of transactions for the average household was 98.27 million. Each one of the variables for all flower types will be compared to the average consumer. In order to do so, the value from the variable will be subtracted from the average household va lue; hence, the comparison between the variables and the average consumer is repres ented as variation from its mean set of characteristics or letting all binary variables equal zero. Figure 6.1 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations From Their Means for Age. 0.00 2.00 4.00 6.00 -2.00 -4.00 -6.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 0.15 -0.05 -0.10 Cut-Flowers Buyer Frequency Deviations From Means Frequency Age under 25 Age 25/39 Age 40/55 Age 55+ 0.00 20.00 40.00 60.00 80.00 100.00 -20.00 -40.00 -60.00 -80.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 144

124 Note again that the penetration may appear small simply because of the dividing by all households instead of the households in a particular demographic group. The first age group, households with 25 years or less, has lower than average market penetration, buyer frequency and, hence, the number of tran sactions. If this age group and buyers with 55 years of age or more are used, which is the age group with the highest overall number of transactions, then one can calculate the pr oportion of the transactions due to frequency of purchasing capturing the whole age effect. About 2.47 percent of the total number of transactions is due to freque ncy of buying in response to cha nges of the whole age effect. Stated differently, almost all of the transac tion gains across age is attributed to market penetration, not frequency of buying. Figure 6.2 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Gender. 0.00 2.00 4.00 6.00 -2.00 -4.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.01 0.02 0.03 0.04 -0.01 -0.02 -0.03 Cut-Flowers Buyer Frequency Deviations From Means Frequency MaleFemale 0.00 20.00 40.00 60.00 80.00 -20.00 -40.00 -60.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 145

125 For the case of gender and purpose there ar e only two categories for each variable; then when calculating the change in frequency one would take the high number of transactions and the low. Females have highe r market penetration values, frequency of transaction and number of transaction than ma les. Keeping the rest of the variables at their average value, if only the male variab le changes the resulting market penetration and frequency of transactions will be lowe r than that for the average consumer. Thus resulting in a lower than average number of transactions as shown in Figure 6.2. The opposite happens if the variable changing is females, and the re st kept constant at their mean values. In order to capture the whole gender effect, equation 6.4 will be used to calculate the change in frequency, using female s as the higher number of transactions and males as the lower number of transactions. The result yielded that 2. 3 percent of changes in gender effects are due to fr equency of transacti ons; again, frequenc y has a very small impact on the total number of transactions. If the purpose of the purchase was for self -use, then the simulations showed that market penetration was lower than the av erage; however, the fr equency of buying was higher than the averag e. This means that in the ca se of households who purchased flowers for self-use, if only th at changed, then the market penetration was lower than average but they were making more transactio ns per period. Households that purchased flowers for gifts had a higher than average market penetration and lower than average frequency of purchasing. From the two va riables, households buying flowers for gifts made the higher number of transactions. From the whole purpose effect variations in the number of transactions of only 1.6 percent were due to the frequency of buying. Figure 6.3 is a graphical representation of market penetration, buyer frequency and number of

PAGE 146

126 transactions for purpose. Interestingly, this is one of the few cases where the penetration and frequency moved in opposite directions. Figure 6.3 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations From Their Means for Purpose. For the case of income, the changes in ma rket penetration, frequency of buying and numbers of transactions were not consistent in their impacts, as shown in Figure 6.4. In order to capture variations in the whole income effect, the income category with the highest number of transactions will be sele cted, $75,000 or more; and the category with the lowest number of transactions, $50,000 to $75,000. After implementing this formula to calculate the delta frequenc y a value of 8.14 percent for the whole income effect was obtained. About 8.14 percent of the total numbe r of transactions is due to frequency of buying in response to changes of the whole income effect. Unlike the previous two 0.00 2.00 4.00 6.00 -2.00 -4.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.01 0.02 0.03 0.04 -0.01 -0.02 -0.03 Cut-Flowers Buyer Frequency Deviations From Means Frequency SelfGift 0.00 20.00 40.00 60.00 80.00 -20.00 -40.00 -60.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 147

127 figures, frequency is relatively more importa nt across income compared with age and purposes for buying. Figure 6.4 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. In terms of seasonality, Figure 6.5 s hows that the market penetration, buyer frequency and number of transactions are high er than average during the period from February to May. These higher than the averag e monthly values may be attributed to the special calendar occasions that occur duri ng this period (e.g., ValentineÂ’s Day and MotherÂ’s Day). Frequency of buying is also higher than average in June, August and December. Overall the month with the highest number of transactions is February, while the least number of transactions occur in June, October and November, the three months 0.00 1.00 2.00 3.00 -1.00 -2.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 Cut-Flowers Buyer Frequency Deviations From Means Frequency Income under 25 Income 25/50 Income 50/75 Income 75+ 0.00 10.00 20.00 30.00 40.00 -10.00 -20.00 -30.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 148

128 sharing nearly the same lower number of transactions. About one percent of the seasonal changes in cut-flowersÂ’ total number of tr ansactions is due to frequency of buying. Figure 6.5 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. Regional behavior of market penetra tion, buyer frequency and number of transactions are presented in Figure 6. 6. The region with the highest number of transactions is Middle Atlantic, while the regi on with the lowest number of transactions is East South Central. About 3.7 percent of the regional changes in cut-flowersÂ’ total number of transactions is due to fre quency of buying. Figure 6.7 represents the percentage of the transactions that is attri buted to frequency of buying cut-flowers for all 0.00 1.00 2.00 3.00 4.00 5.00 -1.00 -2.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 Cut-Flowers Buyer Frequency Deviations From Means Frequency January February March April May June July August September October November December 0.00 20.00 40.00 60.00 80.00 -20.00 -40.00 -60.00 -80.00 -100.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 149

129 variables. If one wants to obtain the percen tage of the transactions due to market penetration we can easily calculate that by subtracting 100delta frequency. Figure 6.6 Cut-Flowers Market Penetra tion, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. As shown in Figure 6.7 the proportion of th e variable changes in the number of transactions corresponding to fr equency of buying for cut-flower s is low. In other words, the change in the variableÂ’s number of trans actions is due in a larger proportion to an increase in the number of buyers. The number of transactions for cutflowers is affected 0.00 2.00 4.00 -2.00 -4.00 Cut-Flowers Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 -0.06 -0.08 Cut-Flowers Buyer Frequency Deviations From Means Frequency New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 20.00 40.00 60.00 80.00 -20.00 -40.00 -60.00 Cut-Flowers Transactions Deviations From Means Transactions

PAGE 150

130 the most by attracting new buyers into the mark et. Therefore, all marketing strategies for cut-flowers need to focus primarily on progr ams that attempt to stimulate non-buyers to become buyers of flowers. Income effects a nd regional effects are the two variables that have the highest percentage of the total num ber of transactions due to frequency of buying. The total number of transactions for seasonal effects on cutflowers is affected almost entirely by the entry of new buyers into the market. The period where the market penetration is at its highest level is from February to May, corresponding to special purchasing occasions. Figure 6.7 Ranges And Percentages of Variable Changes Affecting Transactions Due to Frequency of Buying for Cut-Flowers. The ranges shown in Figure 6.7 are the values of the highest and lowest number of transactions for each variable presented as devi ations from their means. The variable with the largest gap on the number of transactions from the highest to the lowest is age, followed by seasonality and purpose. Income is the variable with the overall lower range on the number of transactions. 2.47 1.02 1.60 3.69 2.30 8.14 Age Seasonality Purpose Region Gender Income Percent of Transactions Due to Frequency 0.002.004.006.008.0010.00 Age Seasonality Purpose Regions Gender Income Range of Transactions From Means0.00 20.00 40.00 60.00 80.00 100.00 -20.00 -40.00 -60.00 -80.00 -100.00 Age Seasonality Purpose Regions Gender Income Range of Transactions0.00 20.00 40.00 60.00 80.00 100.00 120.00 140.00 160.00

PAGE 151

131 Simulations For Flowering Plants And Greens The average household had a market pe netration value of 0.00055 and a buyer frequency value of 1.23919. The calculated nu mber of transactions for the average household was 76.08 million. As was computed fo r cut-flowers, each one of the variables for plants will be compared to the averag e consumer by taking the value from each variable and subtracting the average househol d value. Hence, the comparison between the variables and the average consumer is represen ted as changes from its mean, as shown in Figure 6.8 for age. Figure 6.8 Plants Market Penetration, Buye r Frequency and Number of Transactions Deviations From Their Means for Age. The first age group, households with 25 years or less, has lower than average market penetration, buyer frequency and numb er of transactions. If one uses this age 0.00 2.00 4.00 6.00 8.00 -2.00 -4.00 -6.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 -0.02 -0.04 -0.06 Plants Buyer Frequency Deviations From Means Frequency Age under 25 Age 25/39 Age 40/55 Age 55+ 0.00 50.00 100.00 150.00 -50.00 -100.00 Plants Transactions Deviations From Means Transactions

PAGE 152

132 group and buyers with 55 years of age or more, which is the age group with the highest overall number of transactions, then one can calculate the proportion of the transactions due to frequency of purchasing capturing the whole age effect. About a half percent of the total number of transactions is due to frequency of buying in response to changes across the age levels. The age response for plants almost mirrors that of cut-flowers initially shown in Figure 6.1. Figure 6.9 Plants Market Penetration, Buye r Frequency and Number of Transactions Deviations From Their Means for Gender. For both gender and purpose there are only two categories for each variable when calculating the differences in transactions Females have higher market penetration 0.00 2.00 4.00 6.00 8.00 -2.00 -4.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.01 0.02 0.03 0.04 -0.01 -0.02 -0.03 Plants Buyer Frequency Deviations From Means Frequency MaleFemale 0.00 20.00 40.00 60.00 80.00 100.00 -20.00 -40.00 -60.00 Plants Transactions Deviations From Means Transactions

PAGE 153

133 values, frequency of transaction and number of transaction than males. Keeping the rest of variables at their average value, if only th e male variable changed, the resulting market penetration and frequency of transactions will be lower than that for the average consumer, thus resulting in a lower than average number of transactions. The opposite happens if the variable that changes is female s, and the rest kept constant at their mean values. Using equation 6.4 with M representi ng females and m being males, the results show that 0.85 percent of transaction cha nges between males and females are due to frequency of transactions for plants. Again, the frequency of buying contributes very little to the changes in total demand. If the purpose of the purchase was for self -use, then the simulations showed that market penetration and the frequency of buyi ng and number of transactions were higher than the average for flowering plants. Th is means that house holds who purchased flowering plants for self-use if only that ch anged then the market penetration was higher than average and they were making more tr ansactions per period. Households that purchased plants for gifts have a lower ma rket penetration and lower than average frequency of purchasing. Consequently, from the two purpose levels households buying plants for self-use made the higher numbe r of transactions. From the whole purpose effect of variations in the number of tr ansactions, about 26.5 percent was due to the frequency of buying. Again, market penetratio n contributes most to the changes in total demand. Most plant buyers purchase them for se lf-use, and they make more transactions per month than the average household.

PAGE 154

134 Figure 6.10 Plants Market Pene tration, Buyer Frequency a nd Number of Transactions Deviations From Their Means for Purpose. Across incomes, changes in market pene tration, frequency of buying and numbers of transactions were not linear, as shown in Figure 6.11. In or der to capture variations in the whole income effect, the income categor y with the highest number of transactions will be selected, $75,000 or more, and th e category with the lowest number of transactions, $50,000 to $75,000. After impl ementing the formula to calculate the importance of frequency of buying, one concl udes that about five percent of the total number of transactions is at tributed to frequency of buying in response to changes in across the income levels. 0.00 0.50 1.00 -0.50 -1.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 0.15 -0.05 -0.10 Plants Buyer Frequency Deviations From Means Frequency SelfGift 0.00 10.00 20.00 30.00 -10.00 -20.00 Plants Transactions Deviations From Means Transactions

PAGE 155

135 Figure 6.11 Plants Market Pene tration, Buyer Frequency a nd Number of Transactions Deviations From Their Means for Income. In terms of seasonality, the market pene tration, buyer frequency and number of transactions are higher than average duri ng the period from March to May, and in December as shown in Figure 6.12.. This highe r than the average monthly values, may be attributed to the special cale ndar occasions that occur durin g this period, MotherÂ’s Day and Christmas. Overall the month with the hi ghest number of transactions is December, while the least number of tran sactions occur in August. About 3.53 percent of the seasonal changes in plantsÂ’ total number of tr ansactions are due to frequency of buying. 0.00 1.00 2.00 -1.00 -2.00 -3.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.01 0.02 0.03 0.04 0.05 -0.01 -0.02 -0.03 Plants Buyer Frequency Deviations From Means Frequency Income under 25 Income 25/50 Income 50/75 Income 75+ 0.00 10.00 20.00 30.00 -10.00 -20.00 -30.00 -40.00 Plants Transactions Deviations From Means Transactions

PAGE 156

136 Figure 6.12 Plants Market Pene tration, Buyer Frequency a nd Number of Transactions Deviations From Their Means for Income. Regional behavior of market penetra tion, buyer frequency and number of transactions are presented in Figure 6. 13. The region with the highest number of transactions is Pacific, while the region with the lowest number of transactions is East South Central. About 5.66 perc ent of the regional changes in plantsÂ’ total number of transactions are due to frequency of buying. Figure 6.14 represents the percen tage of the transactions th at is due to frequency of buying for all variables for plants. 0.00 2.00 4.00 6.00 -2.00 -4.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 0.08 0.10 -0.02 -0.04 Plants Buyer Frequency Deviations From Means Frequency January February March April May June July August September October November December 0.00 20.00 40.00 60.00 80.00 100.00 -20.00 -40.00 Plants Transactions Deviations From Means Transactions

PAGE 157

137 Figure 6.13 Plants Market Pene tration, Buyer Frequency a nd Number of Transactions Deviations From Their Means for Income. As shown in Figure 6.14, the proportion of the variable changes in the number of transactions corresponding to fr equency of buying for plants is low. In other words, the change in the variableÂ’s number of transactions is due in a large proportion to an increase in the number of buyers. The numbe r of transactions for plants is affected the most by attracting new buyers into the market. Marke ting strategies for plants need to focus primarily on programs that attempt to stimul ate non-buyers to become buyers of plants. 0.00 1.00 2.00 3.00 -1.00 -2.00 Plants Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 -0.06 Plants Buyer Frequency Deviations From Means Frequency New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 10.00 20.00 30.00 40.00 -10.00 -20.00 -30.00 Plants Transactions Deviations From Means Transactions

PAGE 158

138 Purpose effects and regional eff ects are the two variables that have the highest percentage of the total number of transactions due to frequency of buying. The total number of transactions for both age and gender effects for plants are affected almost entirely by the entry of new buyers into the ma rket. Income effect has about five percent of the total number of transactions due to frequency of purchasing, and its ma rket penetration is higher for the highest income group ($75,000 or more). Figure 6.14 Ranges And Percentages of Vari able Changes Affecting the Number of Transactions Due to Frequency of Buying for Plants. The ranges shown in Figure 6.14 are the valu es of the highest a nd lowest number of transactions for each variable presented as devi ations from their means. The variable with the largest gap on the number of transactions from the highest to the lowest is age, followed by gender and seasonality. Purpose is th e variable with the overall lowest range in the number of transactions, yet the importanc e of frequency of buying is greater for the purpose of buying. 0.50 0.85 3.53 5.66 4.99 26.48 Age Gender Seasonality Region Income Purpose Percent of Transactions Due to Frequency 0.00 5.00 10.00 15.00 20.00 25.00 30.00 Age Gender Seasonality Region Income Purpose Range of Transactions From Means0.00 50.00 100.00 150.00 -50.00 -100.00 Age Gender Seasonality Region Income Purpose Range of Transactions0.0050.00100.00150.00200.00

PAGE 159

139 Simulations For Dry/ Artificial Flowers The average household had a market pe netration value of 0.00013 and a buyer frequency value of 1.48445. The calculated nu mber of transactions for the average household was 20.98 million. Each one of th e variables for dry/artificial will be compared to the average consumer as shown in Figure 6.15 for age. Note that the total number of transactions for this group is consid erably less that for cut-flowers and flowering plants. Figure 6.15 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Age. 0.00 1.00 2.00 3.00 -1.00 -2.00 Dry/Artificial Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 0.08 -0.02 -0.04 -0.06 -0.08 Dry/Artificial Buyer Frequency Deviations From Means Frequency Age under 25 Age 25/39 Age 40/55 Age 55+ 0.00 10.00 20.00 30.00 40.00 50.00 -10.00 -20.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 160

140 The first age group, households under 25 years, have lower market penetration values than the average households, lower frequency of transactions than the average and lower number of transactions than average. The age group with the highest number of transactions is 55 years or more, while age under 25 is the age gr oup with the lowest number of transactions. About 0.32 percent of the total number of transactions is due to frequency of buying in response to changes over the age levels. Again, most of the change in total demand for dry/ artificial flowers for the age variable is attributed to market penetration. Figure 6.16 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Gender. 0.00 1.00 2.00 3.00 4.00 -1.00 -2.00 Dry/Artificial Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 0.15 0.20 -0.05 -0.10 -0.15 Dry/Artificial Buyer Frequency Deviations From Means Frequency MaleFemale 0.00 20.00 40.00 60.00 80.00 -20.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 161

141 For gender, females have higher mark et penetration values, frequency of transaction and number of transaction than male s. Keeping the rest of variables at their average value, if only the male variable ch anged the resulting market penetration and frequency of transactions will be lower than the average household, resulting in a lower than average number of transactions. From equation 6.4, approximately 0.87 percent of changes in gender effects are due to freque ncy of transactions, again a very small number. If the purpose of the purchase was for self -use, then the simulations showed that market penetration was higher than the aver age, and the frequency of buying was also higher than the average. This means that hous eholds who purchased dry/artificial flowers for self-use if only that changed then the ma rket penetration was higher than average and they were making more transactions per peri od. Households that purchased flowers for gifts, had a lower than average market penetration and lower than average frequency of purchasing. From the two variables, house holds buying flowers for self-use made the highest number of transactions and households buying flowers for gifts made the lowest number of transactions. From the whole purpo se effect, 7 percent of the change in the number of transactions was attr ibuted to the frequency of buyi ng. Most of the variation in transactions of dry/artificial for purpose is due to market penetration. Figure 6.17 is a graphical representation of market pe netration, buyer frequency and number of transactions for purpose.

PAGE 162

142 Figure 6.17 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Purpose. For the case of income, changes in mark et penetration, frequency of buying and numbers of transactions were not linear, as shown in Figure 6.18. In order to capture variations in the whole income effect, one will select the inco me category with the highest number of transacti ons, under $25,000, and the category with the lowest number of transactions, $50,000 to $75,000. After im plementing the formula to calculate the relative frequency, a value of 3.48 percent for the whole in come effect was obtained. About 3.48 percent of the tota l number of transactions is due to frequency of buying in response to changes of the income effect. 0.00 1.00 2.00 3.00 4.00 -1.00 -2.00 -3.00 -4.00 Dry/Artificial Market Penetration Deviation From Means 1E-5 Penetration 0.00 0.01 0.02 0.03 0.04 -0.01 -0.02 -0.03 -0.04 Dry/Artificial Buyer Frequency Deviations From Means Frequency SelfGift 0.00 2.00 4.00 6.00 -2.00 -4.00 -6.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 163

143 Figure 6.18 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Income. Figure 6.19 shows that the market penetra tion, frequency of buying and number of transactions were higher durin g the period from February to May, in August and in the period November-December. The highest numb er of transactions occured during the period from February to May, with May having the largest number of transactions. This higher than the average monthly values may be attributed to the special calendar occasions that occur during this period, Va lentineÂ’s Day and MotherÂ’s Day. The month with the highest number of transactions is May, while the least number of transactions 0.00 2.00 4.00 6.00 -2.00 -4.00 -6.00 -8.00 Dry/Artificial Market Penetration Deviation From Means 1E-5 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 -0.06 Dry/Artificial Buyer Frequency Deviations From Means Frequency Income under 25 Income 25/50 Income 50/75 Income 75+ 0.00 5.00 10.00 -5.00 -10.00 -15.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 164

144 occur in July. About 10.23 percent of the s easonal changes in dry/ar tificial total number of transactions are due to frequency of buying. Figure 6.19 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations From Their Means for Seasonality. Regional behavior of market penetra tion, buyer frequency and number of transactions are presented in Figure 6. 20. The region with the highest number of transactions is South Atlantic, while the regi on with the lowest number of transactions is New England. About 2.51 percent of the region al changes in dry/artif icial total number of transactions is due to frequency of buying. 0.00 2.00 4.00 6.00 8.00 -2.00 -4.00 -6.00 Dry/Artificial Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 0.15 -0.05 -0.10 Dry/Artificial Buyer Frequency Deviations From Means Frequency January February March April May June July August September October November December 0.00 5.00 10.00 15.00 20.00 -5.00 -10.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 165

145 Figure 6.21 represents the percen tage of the transactions th at is due to frequency of buying for all variables for dry/artificial. Figure 6.20 Dry/Artificial Ma rket Penetration, Buyer Frequency and Number of Transactions Deviations Fr om Their Means for Regions. As shown in Figure 6.21 the proportion of the variable changes in the number of transactions correspond ing to frequency of buying for dr y/artificial are low. In other words, the change in the variableÂ’s number of transactions is due in a larger proportion to an increase in the number of buyers. The number of transactions for dry/ artificial is most 0.00 0.50 1.00 -0.50 -1.00 Dry/Artificial Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 0.15 -0.05 -0.10 Dry/Artificial Buyer Frequency Deviations From Means Frequency New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 5.00 10.00 15.00 20.00 -5.00 -10.00 -15.00 Dry/Artificial Transactions Deviations From Means Transactions

PAGE 166

146 affected by attracting new buyers into the ma rket. Seasonal and purpose effects are the two variables that have the highest percentage of the total number of transactions that are due to frequency of buying. The total number of transactions for age and gender effects on dry/artificial is affected almost entirely by the entry of new buyers into the market. The period where the frequency of buying is at its highest level is from February to May, corresponding to special purchasing occasions. Figure 6.21 Ranges And Percentages of Vari able Changes Affecting the Number of Transactions Due to Frequency of Buying for Dry. The ranges shown in Figure 6.21 are the valu es of the highest a nd lowest number of transactions for each variable presented as devi ations from their means. The variable with the largest gap on the number of transactions from the highest to the lowest is gender, followed by age and region. Pu rpose is the variable with the overall lower transaction range. 0.87 0.32 2.51 10.23 3.48 7.04 Gender Age Region Seasonality Income Purpose Percent of Transactions Due to Frequency 0.00 2.00 4.00 6.00 8.00 10.00 12.00 Gender Age Region Seasonality Income Purpose Range of Transactions From Means0.00 20.00 40.00 60.00 80.00 -20.00 Gender Age Region Seasonality Income Purpose Range of Transactions0.0020.0040.0060.0080.00100.00

PAGE 167

147 Simulations For Outdoor The average household had a market pe netration value of 0.00051 and a buyer frequency value of 1.52204. The calculated nu mber of transactions for the average household was 85.75 million. Each one of th e variables for outdoor flowers will be compared to the average consumer. The comparison between the variables and the average consumer is represented as variati on from its mean, as shown in Figure 6.22 for age. Figure 6.22 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Age. 0.00 2.00 4.00 6.00 8.00 10.00 -2.00 -4.00 -6.00 Outdoor Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.05 0.10 -0.05 -0.10 -0.15 Outdoor Buyer Frequency Deviations From Means Frequency Age under 25 Age 25/39 Age 40/55 Age 55+ 0.00 50.00 100.00 150.00 200.00 -50.00 -100.00 Outdoor Transactions Deviations From Means Transactions

PAGE 168

148 The first age group, households with 25 years or less, has the lowest market penetration, buyer frequency and number of tr ansactions. If one use this age group and buyers with 55 years of age or more, which is the age group with the highest overall number of transactions, then one can calculate the proportion of the transactions due to frequency of purchasing capturing the whole ag e effect. About 0.43 percent of the total number of transactions for outdoor flowers is due to frequency of buying in response to changes of the whole age effect. Figure 6.23 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Gender. 0.00 2.00 4.00 6.00 8.00 -2.00 -4.00 Outdoor Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 0.08 -0.02 -0.04 -0.06 Outdoor Buyer Frequency Deviations From Means Frequency MaleFemale 0.00 50.00 100.00 150.00 -50.00 -100.00 Outdoor Transactions Deviations From Means Transactions

PAGE 169

149 For the case of gender and purpose there ar e only two categories for each variable; when calculating the change in frequency one would take the high number of transactions and the low. Females have higher market pene tration values, frequency of transaction and number of transaction than males. Keeping th e rest of variables at their average value, male changes will result in a lower market penetration and higher frequency of transactions than that for the average cons umer. The number of transactions was lower than average number of transactions. In order to capture the whole gender effect, equation 6.4 will be used to calculate the change in frequency, using females as the higher number of transactions and males as th e lower number of transactions. The result yielded that 1.47 percent of changes in gender effects are due to frequency of transactions. If the purpose of the purchase was for self -use, then the simulations showed that market penetration, buyer frequency of pur chasing and number of transactions were higher than the average .For households that purchased outdoor flowers for gifts, the values were lower than average market penetr ation and higher than average frequency of purchasing. The number of transactions wa s higher than average when purchases were made for self-use and lower than average when purchases were made with the purpose of gifts. From the two variables, households buying outdoor flowers fo r self-use made the higher number of transactions. From the whole purpose effect, about 7 percent was attributed to the frequency of buying in the case of outdoor-flowers. Figure 6.24 is a graphical representation of market pe netration, buyer frequency and number of transactions for purpos e for outdoor flowers.

PAGE 170

150 Figure 6.24 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Purpose. For the case of income, changes in mark et penetration, frequency of buying and numbers of transactions were not linear, as shown in Figure 6.25. In order to capture variations in the whole income effect, one will select the inco me category with the highest number of transactions, $75,000 or more, and the categor y with the lowest number of transactions, $50,000 to $75,000. Af ter implementing the formula to calculate the relative frequency, a value of 6.05 percent for the whole income effect was obtained. About 6.05 percent of the tota l number of transactions is due to frequency of buying in response to changes of the income effect. 0.00 2.00 4.00 6.00 -2.00 -4.00 Outdoor Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.10 0.20 0.30 -0.10 -0.20 -0.30 Outdoor Buyer Frequency Deviations From Means Frequency SelfGift 0.00 50.00 100.00 150.00 -50.00 -100.00 Outdoor Transactions Deviations From Means Transactions

PAGE 171

151 Figure 6.25 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Income. Figure 6.26 shows that the market pene tration, buyer frequency and number of transactions are higher than average during the period from March to June. These higher than the average monthly values may be attribut ed to the special fact that the season for outdoor plants starts on this dates, or to special calendar occasions Frequency of buying is also higher than average during the same period. The month with the highest number of transactions is May, while the least number of transactio ns occur in December. About one percent of the seasonal ch anges in outdoor flow ersÂ’ total number of transactions are 0.00 0.50 1.00 1.50 2.00 -0.50 -1.00 -1.50 -2.00 Outdoor Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 Outdoor Buyer Frequency Deviations From Means Frequency Income under 25 Income 25/50 Income 50/75 Income 75+ 0.00 10.00 20.00 30.00 -10.00 -20.00 -30.00 Outdoor Transactions Deviations From Means Transactions

PAGE 172

152 due to frequency of buying. Clearly, there are major seasonal differences among the form flower types. Figure 6.26 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Income. Regional behavior of market penetra tion, buyer frequency and number of transactions are presented in Figure 6. 27. The region with the highest number of transactions for outdoor is Pacific, while the region with the lowest number of transactions is Mountain. A bout 3.58 percent of the regiona l changes in out-doorÂ’s total number of transactions is due to frequency of buying. 0.00 0.50 1.00 1.50 2.00 2.50 3.00 -0.50 Outdoor Market Penetration Deviation From Means 1E-3 Penetration 0.00 0.20 0.40 0.60 -0.20 -0.40 Outdoor Buyer Frequency Deviations From Means Frequency January February March April May June July August September October November December 0.00 200.00 400.00 600.00 800.00 -200.00 Outdoor Transactions Deviations From Means Transactions

PAGE 173

153 Figure 6.28 represents the percen tage of the transactions th at is due to frequency of buying for all variables for out-door flowers. Figure 6.27 Outdoor Market Pene tration, Buyer Frequency an d Number of Transactions Deviations From Their Means for Income. As shown in Figure 6.28 the proportion of the variable changes in the number of transactions corresponding to fr equency of buying for outdoor flowers is low. In other words, the change in the variableÂ’s number of transactions is due in a larger proportion to an increase in the number of buyers. The numbe r of transactions for outdoor flowers is 0.00 1.00 2.00 3.00 4.00 -1.00 -2.00 -3.00 Outdoor Market Penetration Deviation From Means 1E-4 Penetration 0.00 0.02 0.04 0.06 -0.02 -0.04 -0.06 Outdoor Buyer Frequency Deviations From Means Frequency New England Middle Atlantic East North Central West North Central South Atlantic East South Central West South Central Mountain Pacific 0.00 20.00 40.00 60.00 80.00 -20.00 -40.00 Outdoor Transactions Deviations From Means Transactions

PAGE 174

154 most affected by attracting new buyers into the market. Purpose effects and income effects are the two variables that have the highest percentage of the total number of transactions that are due to frequency of buying. The total number of transactions for seasonal effects on outdoor flowers is affected almost entirely by the entry of new buyers into the market. The period where the market penetration is at its highest level is from April to June, corresponding to the be ginning of the outdoor flowersÂ’ season. Figure 6.28 Ranges And Percentages of Vari able Changes Affecting the Number of Transactions Due to Frequency of Buying for Outdoor. The ranges shown in Figure 6.28 are the valu es of the highest a nd lowest number of transactions for each variable presented as devi ations from their means. The variable with the largest gap on the number of transactions from the highest to the lowest is seasonality, followed by age and gender. Income is the vari able with the overall lower range on the number of transactions. 0.96 0.43 1.47 7.04 3.58 6.05 Seasonality Age Gender Purpose Region Income Percent of Transactions Due to Frequency 0.002.004.006.008.00 Seasonality Age Gender Purpose Region Income Range of Transactions From Means0.00 200.00 400.00 600.00 800.00 -200.00 Seasonality Age Gender Purpose Region Income Range of Transactions0.00200.00400.00600.00800.00

PAGE 175

155 Chapter Summary The main objective of the sensitivity anal ysis was to obtain the expected value of market penetration and buyer fr equency for all flower types. As stated in the introductory section, from the simulation analysis the e xpected value of the dependent variables was obtained using equations 4.11-4.15 descri bed in the methodology chapter. Results fluctuate from one flower t ype to another and also depe nd on the variable that was changing. One of the most important overall ob jectives of this research project was to separate the demand for flowers into the mark et penetration component from that of the frequency of buying. Most of th e number of transactions for all flowers took place by the entry of new buyers rather than the frequency of buying; however, when analyzing each variable individually, this percentage differe d across flower types. Figure 6.29 presents a summary of the percentage of the number of transactions that is due to frequency of buying for all flower types. Figure 6.29 Percentage of Tr ansactions Due to Frequency of Buying For All Flower Types. 8.14 3.69 2.47 2.30 1.60 1.02 Income Region Age Gender Purpose Seasonality Percent of Transactions Due to Frequency Cut-Flowers 0.002.004.006.008.0010.00 4.99 5.66 0.50 0.85 26.48 3.53 Income Region Age Gender Purpose Seasonality Percent of Transactions Due to Frequency Plants 0.00 5.00 10.00 15.00 20.00 25.00 30.00 3.48 2.51 0.32 0.87 7.04 10.23 Income Region Age Gender Purpose Seasonality Percent of Transactions Due to Frequency Dry/Artificial 0.00 2.00 4.00 6.00 8.00 10.00 12.00 6.05 3.58 0.43 1.47 7.04 0.96 Income Region Age Gender Purpose Seasonality Percent of Transactions Due to Frequency Outdoor 0.002.004.006.008.00

PAGE 176

156 For cut-flowers the variable that had the hi ghest percentage of transactions due to frequency of buying was income, followe d by region, age, gender, purpose and seasonality. This result was not the same for the rest of the flower types as shown in Figure 6.29. These findings can be used to design marketing strategies that target specific household groups for a specific fl ower type in an attempt to increase the overall total demand for flowers. Overall for each flower type ranging from cut-flowers to outdoor flowers, entry of buyers or market penetration dominated in its contribution to th e changes in demand. Frequency of buying showed the largest impact for flowering plants and greens. Yet even with these plants, frequency still accounted for less than 27 percent of changes in demand. The result is, however, quite consiste nt in terms of the importance of market penetration.

PAGE 177

157 CHAPTER 7 SUMMARY AND CONCLUSIONS This chapter presents a brief summary of the research. In the first section, background information of flower demand w ill be introduced, followed by some trends of the flower industry in the United States. Then, the objectives of this study will be briefly discussed, including empirical results and what the implications of these results are for the demand of flowers in the Unite d States. The last section includes the limitations of this study, as well as recommenda tions of the direction for future research in flower demand. Summary and Conclusions Consumption behavior has always been of great importance and a topic of focus for researchers. The consumption of goods takes place because of the satisfaction that the goods or services provide. The consumption of traditional agricultural food products depends on the characteristic s of the product or attributes that can be measured or quantified. In contrast to food products, ma ny nonfood products are consumed because of their aesthetic value. Flowers are purchased for various reasons such as expression of love or friendship, a way to express thankfulness or appreciat ion, beautification purposes for self or gift. The attributes of flowers, or more generally non-food products cannot be quantified; therefore the satisfaction gained from the consumption of these goods is closely related to the objective of the purchas e. This also implies that the demand for these products can be influenced by the char acteristics or preferen ces of the buyers and the reason for buying the products. This fact can be viewed during special calendar

PAGE 178

158 occasions (i.e., MotherÂ’s Day, ValentineÂ’s, et c), where the consumpti on of floral products is substantially higher compared to non-calendar occasions. Demand for all products depends on the charac teristics or attributes of the products. For most food products the prevai ling characteristic is to sati sfy nutritional needs. Even though flowers are not essential for survival they possess other characteristics of importance to food products that influence the buying decision. Because flowers are not essential for survival there is a portion of the population composed by non-buyers or infrequent buyers. Therefore there is a cons iderable gap for the decision of buying or not, and this decision is based upon the demographics of the population and the occasions and periods. Understanding how consumers make choices whether to buy or not and the perceptions of the characterist ics of the products are essent ial to understanding the flower demand. There are three groups of factors that affect the demand for floral products: external, controlled, and seasonal factors. Ex ternal factors of dema nd include inflation, wages, prices, unemployment rate, demogra phic factors and other economic variables. Controlled factors of demand may be used to change perceptions and awareness with the use of promotions, product developments a nd innovations. Seasonal factors also affect the demand for flowers. There are certain calendar occasions where the demand for flowers is higher compared to other non-cal endar occasions. The most common calendar occasion dates are MotherÂ’s Day and ValentineÂ’s Day. In order to analyze the demand for flowers two types of analysis were made for cutflowers, potted flowering plan ts, dry/artificial a nd outdoor flowers. First, market penetration and second buyer frequency. Becau se flowers are non-essential for survival,

PAGE 179

159 in a typical month the percentage of the popul ation that buys flower s is less than five percent. From this fact arises the need to understand how consumers make the choice to purchase or not and what the factors are that influence their purchasing decisions. After determining the factors that affect their pur chase behavior, simulation analysis was used to develop specific programs to increase th e entry of new consumers. Once a person becomes a consumer of flowers, the remaini ng question is the freque ncy of buying. It is also of great importance to understand the fa ctors that influence th e purchasing decisions among consumers of flowers in an attemp t to increase the total consumption. Even though fresh cut flowers, potted flow ering plants, and dryartificial flowers are fundamentally different and substitutable to some degree, there are certain similarities in their attributes between these products if analyzed in terms of the purpose of use. They can be used to express love, thanks, re flect emotions, project beauty, and show environmental concerns. Consumer expend iture patterns may change between these products even though they are physically differe nt. These consumer patterns are affected by many factors, including income, purpose of use, occasions, information and perceptions and sources for purchases. The level of consumer e xpenditures depends on three basic components: market penetration, frequency of transactions among buyers and prices. Demand analyses for fl oral products differ among othe r agricultural commodities in the sense that for other agricultural commodities the quantity consumed is used directly in the analysis. In the specific cas e of flowers, a consumer may purchase a single stem rose or an arrangement. Therefor e demand studies on flow ers generally replace quantity observed by the number of transac tions given on a defined period of time.

PAGE 180

160 Given the reasons why consumers buy flowers and because flowers are nonessential for survival, one can observe that the consumption of floral products had significant fluctuations over time. These fl uctuations resulted in a higher number of transactions during special calendar occasi ons during the year. The most important calendar occasions were ValentineÂ’s, MotherÂ’s day, Easter/Passove r, Thanksgiving and Christmas/Hanukah. In the last decade, the industry has e xperience many changes including industry programs adopted to increase the total demand for flowers. Brand a nd generic programs have been adopted to entice the demand for floral products. In addition to promotion programs, the development of new technologies, such as the Intern et, has made possible the creation of new sources for buying floral products. The data set for flower purchases from July 1992 to July 2004 was obtained from the American Floral Endowment (AFE) and Ip sos-NPD group. These data was based in a consumer panel of several thousand households who reported their purchases of floral products in the US. The data set is organi zed by number of households, expenditures, transactions and buyers. Market penetration was defi ned as number of buyers divided by the number of households. This would result in a market pe netration index between the values of zero and one, where zero means that there are no buyers at all, wh ile a value above zero means that some households with a defined group were buyers. Because both models, market penetration and buyer frequency, have a cluster of observations on the lower limit as shown in Tables 4.1 and 4.2, a model was sele cted that takes into account its asymptotic distribution. The market penetration model has a lower limit at zero, while the buyer

PAGE 181

161 frequency has a lower limit of one, since in order to be defined as a buyer a household must have made by definition at least one tr ansaction per month or more. The model that deals with this type of clustering of the data is the Tobit model. There were two main demand models in this research, each one including a market penetration factor and a buyer frequency. Th e first model had both regions and flower types as dependent variables, and therefore re sulted in one equation for each flower type and region. The second model included region s as independent variables, and therefore there was one equation for each flower type. The objective of having these two different models was to allow two different types of analyses. In the first model, a specific equation can be use to gain insight into the demand for a specific flower type on a specific region, by only containi ng data from that specific region and flower type. This possible was due to the large number of obser vations for each factor analyzed. In the second model, one equation was used for each product form; regions were included as independent dummy variables. This model allo w us to analyze regional changes for each one of the product forms, and allows the fl ower industry to design marketing programs that target specific flower t ypes, regions and demographics. The results from both demand models yielde d similar results. All dummy variables for all models had the average of all cate gories as a base; ther efore the parameter estimates are deviations from the average household. For example, if the parameter estimate for the month of February is positive and significant, it means that the month of February is statistically diffe rent (higher) from the averag e of the twelve-month cycle. Most of the parameter estimates in both mode ls were significant at the 95 % confidence level. The results for flower types and region s differed considerably. Their interpretation

PAGE 182

162 is quite simple, as all of the variables except price are dumm y variables and hence represent deviations from its means. In general it was found that the demand fo r flowers, both market penetration and buyer frequency, depends on demographic ch aracteristics, purpose of the purchase and seasonal factors. The results vary dependi ng on the flower types and regions. For example, for most cut-flowers, it was found that market penetra tion and buyer frequency increased with females purchases for the purpose of self and with the higher ages categories. In general, for seasonality effect s, each month was compared to an average over the twelve-month period. The results indi cated that household decisions to purchase flowers (market penetration) and the numbe r of transactions on a given period (buyer frequency) were highly impacted by calen dar occasions. The only continuous variable was price on the buyer frequenc y models. For all models a nd equations the sign of the parameter estimate for price was negative. This is in accordance with economic theory for normal goods. The rest of the paramete r estimates were obtained from dummy variables and can be interpreted easily as devi ations from its means. The complete set of results is presented in the appendix secti on. The first model has 140 equations, 70 for market penetration and 70 equations for buyer frequency. The second model consists of 14 equations, seven for market penetration an d seven for buyer frequency. Most of the analyses concentrated in the second type models. The simulation analysis is an essential part of this research project. Each simulation procedure measured demand changes by adjusti ng one or more variables relative to the mean value of the rest of all variables in the demand model. The first step in the simulation analysis was to cal culate the market penetration and buyer frequency values

PAGE 183

163 for the average household consumer. Then, bot h market penetration and buyer frequency were calculated with changes in one variable only and the rest of the variables kept constant at the average cons umer. After obtaining the values for market penetration and buyer frequency, it was explaine d what proportion of the total number of transactions was due to the frequency of buying versus the increment in the number of buyers (market penetration). In order to do so, first the market penetration value was used and multiplied by the total number of households to obtain the total number of buyers (BUY). Once the total number of buyers and the frequency of transaction (FRQ) were obtained for the average household and for changes within a sp ecific variable. Then the proportion of the variable attributed to buye r frequency versus market penetration was calculated. For example, for age, the highest and lowest numbe r of transactions was selected, in order to capture the whole variation from age. The proportion of the variable changes in the number of transactions corresponding to frequency of buying for cut-flowers, plants dry/artificial, and outdoor are low, varying from one flower type to the other. In other words, the change in the variableÂ’s number of transactions is due in a la rger proportion to an increas e in the number of buyers. The number of transactions for all flower type s was most affected most by attracting new buyers into the market. In the previous chapters and text above, it is clear that the demand for flowers, ranging from cut-flowers to outdoor flowers, is driven in part by demographics, seasonal occasions, purpose, price and geographical di fferences based on regions in the U.S. Furthermore, the demand response is from both changes in the level of market penetration and frequency of buying with the penetration being the major component in

PAGE 184

164 the demand equation. Important differences in the demand drivers were seen across the four types of flowers, (i.e., cut-flowers, flowering plants and greens, dry/artificial flowers, and outdoor flowers). Also, the driver s influenced both the market penetration and frequency of buying with the level of im portance quite different across the drivers within each flower type. Demand for flowers in all forms is a direct reflection of consumer preferences and differences in the preferences across the population. Measuring this demand and its two components, as was done in this study, is essential to understanding and influencing the longer-term growth and opport unities for marketing flowers in the U.S. We know that unlike many other countries, the percentage of U.S. households buying flowers within a month or so is quite low a nd differs by flower type. Figures 6.7, 6.14, 6.21, and 6.28 provided clear insights into these differences across flower types and the demand drivers. For each sector, the obvious goal would be to move the averag e number of total transactions to higher levels. Much of that could probably be accomplished by addressing the factors to generate transact ion levels below the means as initially illustrated in Figure 6.7. For fresh cut-flowers age and seasonality are the two demand drivers having the greatest potential negative impact with the values below the average level of transactions being nearly equal between these two variable s. Then purpose, regional differences and gender produce similar relative effects on the number of tota l transactions. Furthermore, for each of these variables most of the ch anges above or below the average level are attributed to buyer penetration. From Figure 6. 7 the results point to programs to address the age effect and seasonality negative effect s to have probably the most potential to move the average transaction levels even higher. While the regions, purpose (i.e., gift

PAGE 185

165 versus self) and gender have s lightly lower negative impacts, these three are likely easy to target. Recent programs developed by the Flower Promotion Orga nization, a relative new generic promotion program, currently targ et females to buy flowers for self-use and the promotions are targeted to specific re gions. The regional differences shown in the Tobit models should give guidan ce to better regional targeting to the extent that there is flexibility in the regional selection. Fina lly, targeting income groups appears to have considerably less potential relative to the other demand drivers for fresh cut-flowers. In contrast to flowering pl ants and greens, compared to cut-flowers, age and gender have the potential largest nega tive effects as was shown in Figure 6.14. Hence, programs designed to target age and gender have consid erable potential whereas efforts to address seasonal and regional differences, as well as income and purpose, ha ve far less potential to moving the transaction levels for flowering plants. Interestingly, the role of purpose is extremely small, causing very little variat ion in transactions below the mean. Clearly, targeting those age groups and gender that contribute to the negative side of the transaction equation is suggested with the estimates. Also, for both cut-flowers and plants, developing facilities to deal with ne w buyers is important since additional market penetration is where the major gains can occur. For dry and artificial flowers, age and gender are the two most important targets since some age groups and gender create most of the transactions be low the mean levels (see Figure 6.21). Among all fou r-flower types, gender is most important in relative terms for the dry/artificial flower group. Negative e ffects from regional differences, seasonality, income and purpose are very small and mo st likely have limited payoff in producing larger gains in the number of transactions for the dry and artif icial flower group.

PAGE 186

166 Finally, the outdoor flowers show a profoundl y different response level with most of the variation in the transact ions being attributed to season ality. This obviously reflects much of the spring planting season w ith outdoor flowers. Beyond the seasonal differences, age, gender and purpose on the ne gative side of the equation (i.e., producing values below the average) were reasonably small in relative terms as seen in Figure 6.28. Addressing seasonal patterns is likely the most difficult thing to change since the season demand is closely tied to weather, fixed holidays and seasonal celebrations. Also, the importance of frequency of buying is slightly greater for the outdoor market than for the other flower types. There is probably more substitutability among cut-flowers, plants and dry/artificial flowers compared with the outdoor flowers. While the goals likely differ among the four flower types, there are several generalities that have potentia l for all four. The demand for each flower type was closely tied to the age of the buyer with the transactions increasing with the age of the buyer. Hence, efforts to target the younger market in all flower types should have potential positive benefits in all four groups. For the other classifications, programs targeting specific household attributes shou ld more likely be tailored to the type of flowers (e.g., cut, plant, dry or outdoor) being marketed. In all of the examples included in the previous chapters, there was not consideration of the potential gains from ch anging several factors simultaneously. For example, the gains from both changes in ag e, gender and purpose were not shown. The usefulness of that additional approach would be to show the combined effect on both the positive and negative side and, particularly, the possible benefits if the negative side was someway countered through promotions or other type of marketing efforts.

PAGE 187

167 Finally, the extreme importance of the ma rket penetration versus frequency of buying has considerable implications. New buyers may need additional information and are potentially influenced by the first impression, whether the facilities or quality of the flowers. Buying habits may not be as well esta blished in terms of the types of flowers and what is communicated with different types. He nce, having in-store information to guide potential buyers is more important than with products where the consumer is a frequent repeat buyer. Also, the new buyer may be more receptive to in-store displays and promotional materials. For the outdoor flower s, the informational needs are even more challenging for the new buyer. Store layout, re source materials, and personal assistance are likely more important with the demand ga ins coming mostly from market penetration versus the frequency of buying. What this study shows is the potential for gains (or loses) across demographics and non-demographic variables. The specific mech anisms and cost of achieving those gains were not shown. Obvious mechanisms include targeted media advertising; selected magazine advertising; in-store promoti ons; resource centers; we b-pages and related services; locations in stores; and creative ce nters (especially for self use), etc. Limitations and Direction for Future Research Because of the nature of the data set used on this study, the calculation of the average price per transaction was based in an underestimation of the true average price in the case of market penetration. This occurre d because the data set only had information on buyers, as explained on the methodology chapte r. Therefore price wa s not included as one of the independent variab les for the market penetration models. The fact that the effect of price on market penetration was not included also affected the simulations for

PAGE 188

168 price, since changes in price did not affect ma rket penetration because it was not part of the model. The same methodology used in this study can be used to obtain factors affecting the demand for flowers by specific analysis of de mand on specific cities not regions and can also include other demographic characteristi cs such as ethnical groups, outlet sources, etc. The main focus on this demand analysis was on general flower categories, such as cut-flowers, plants, dry/artif icial, outdoor, etc. Demand studies can be developed on specific flower products, such as roses, or poinsettias for the Christmas period.

PAGE 189

169 APPENDIX A MODEL I RESULTS Table A.1 Market Penetration Model I Re sults for Indoor and Cut-Flowers in New England BETA T-VALUE BETA T-VALUE C -0.00074-8.86628-0.00227-25.56958 DINC2 (25/50) 0.000393.400060.000534.99837 DINC3 (50/75) -0.00137-11.37398-0.00082-7.37256 DINC4 (75+) 0.0016314.036660.0010910.18222 DGEN2 (Female) 0.0026637.525610.0015924.16023 DPUR2 (Gifts) 0.0007310.607870.0013220.20770 DAGE2 (25/39) 0.000221.882660.000050.45095 DAGE3 (40/55) 0.0021518.655880.0018817.68646 DAGE4 (55+) 0.0034329.885670.0024122.64676 DMTH2 (February) 0.001366.248910.001728.80511 DMTH3 (March) 0.001547.073640.001628.21217 DMTH4 (April) 0.001607.384070.000663.26921 DMTH5 (May) 0.0024611.561150.001356.82613 DMTH6 (June) -0.00024-1.06394-0.00043-2.04687 DMTH7 (July) -0.00154-6.57502-0.00061-2.87495 DMTH8 (August) -0.00176-7.41091-0.00084-3.91259 DMTH9 (September) -0.00153-6.51862-0.00104-4.79576 DMTH10(October) -0.00100-4.34813-0.00054-2.55902 DMTH11(November) -0.00038-1.67537-0.00020-0.93753 DMTH12(December) 0.000833.75673-0.00092-4.27228 SIGMA 0.0055788.722160.0048273.53323 MARKET PENETRATION NEW ENGLAND INDOOR CUT-FLOWERS

PAGE 190

170 Table A.2 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in New England BETA T-VALUE BETA T-VALUE C -0.00700-28.88921-0.00284-29.23425 DINC2 (25/50) 0.000191.330450.000655.86602 DINC3 (50/75) -0.00047-3.19928-0.00075-6.52109 DINC4 (75+) 0.001299.614370.000837.46330 DGEN2 (Female) 0.0010712.090890.0014721.48929 DPUR2 (Gifts) 0.0025221.309350.0008813.04868 DAGE2 (25/39) 0.000453.01676-0.00003-0.22540 DAGE3 (40/55) 0.001459.965640.0018116.35072 DAGE4 (55+) 0.0016111.117880.0023321.08971 DMTH2 (February) 0.001134.573080.001617.96983 DMTH3 (March) -0.00007-0.240410.001758.66532 DMTH4 (April) 0.000752.945430.000532.51892 DMTH5 (May) 0.001365.574130.001175.70528 DMTH6 (June) -0.00114-3.66945-0.00018-0.83752 DMTH7 (July) -0.00062-2.15106-0.00043-1.94503 DMTH8 (August) -0.00074-2.52665-0.00064-2.89213 DMTH9 (September) -0.00056-1.94064-0.00100-4.39921 DMTH10(October) -0.00064-2.19878-0.00045-2.02937 DMTH11(November) 0.000471.79999-0.00037-1.67038 DMTH12(December) 0.000622.42826-0.00126-5.48979 SIGMA 0.0041836.020470.0048968.39885 MARKET PENETRATION NEW ENGLAND ARRANGE NON-ARRAN

PAGE 191

171 Table A.3 Market Penetration Model I Resu lts for Plants and Dry/Artificial in New England BETA T-VALUE BETA T-VALUE C -0.00372-34.17316-0.00777-27.52150 DINC2 (25/50) 0.000111.02121-0.00024-1.58180 DINC3 (50/75) -0.00122-10.36813-0.00035-2.27780 DINC4 (75+) 0.0013112.200530.000825.70830 DGEN2 (Female) 0.0024233.400740.0030020.49317 DPUR2 (Gifts) -0.00048-7.50365-0.00086-9.43153 DAGE2 (25/39) 0.000201.769950.000543.31132 DAGE3 (40/55) 0.0016514.850980.001449.03654 DAGE4 (55+) 0.0027825.226050.0019712.36002 DMTH2 (February) 0.000090.43206-0.00048-1.57068 DMTH3 (March) 0.000723.506120.001174.49334 DMTH4 (April) 0.0020310.446610.000742.74113 DMTH5 (May) 0.0023011.937310.000853.18000 DMTH6 (June) 0.000180.85523-0.00040-1.32488 DMTH7 (July) -0.00174-7.28781-0.00125-3.71773 DMTH8 (August) -0.00210-8.55713-0.00046-1.51055 DMTH9 (September) -0.00113-4.94382-0.00047-1.54301 DMTH10(October) -0.00120-5.255290.000632.32124 DMTH11(November) -0.00049-2.260440.000501.81994 DMTH12(December) 0.0024312.55854-0.00017-0.58141 SIGMA 0.0044964.237970.0042134.07353 MARKET PENETRATION NEW ENGLAND PLANTS DRY

PAGE 192

172 Table A.4 Market Penetration Model I Results for Outdoor in New England BETA T-VALUE C -0.00881-40.43104 DINC2 (25/50) 0.000281.46917 DINC3 (50/75) -0.00062-3.20580 DINC4 (75+) 0.0020511.08884 DGEN2 (Female) 0.0033527.96419 DPUR2 (Gifts) -0.00200-17.66305 DAGE2 (25/39) 0.000321.54734 DAGE3 (40/55) 0.0032416.46895 DAGE4 (55+) 0.0051426.33434 DMTH2 (February) -0.00471-9.66805 DMTH3 (March) -0.00094-2.43503 DMTH4 (April) 0.0053917.03714 DMTH5 (May) 0.0114337.74003 DMTH6 (June) 0.0061519.53495 DMTH7 (July) 0.001644.65608 DMTH8 (August) 0.000260.70829 DMTH9 (September) 0.001995.74182 DMTH10(October) -0.00070-1.80099 DMTH11(November) -0.00658-11.74890 DMTH12(December) -0.00631-11.48506 SIGMA 0.0069062.71537 MARKET PENETRATION NEW ENGLAND OUTDOOR

PAGE 193

173 Table A.5 Buyer Frequency Model I Results for Indoor and Cut-Flow ers in New England BETA T-VALUE BETA T-VALUE C 0.084560.230040.756430.71606 DINC2 (25/50) -0.13207-2.02002-0.39922-3.04646 DINC3 (50/75) 0.055260.962900.189672.03357 DINC4 (75+) 0.057791.289470.024350.32154 DGEN2 (Female) 0.096630.84455-0.11944-0.55990 DPUR2 (Gifts) -0.29337-5.43488-0.50877-2.75102 DAGE2 (25/39) -0.06564-1.10857-0.18613-2.27723 DAGE3 (40/55) 0.252641.80067-0.02833-0.08770 DAGE4 (55+) 0.159591.10004-0.02583-0.08146 DMTH2 (February) 0.070930.76244-0.04848-0.23944 DMTH3 (March) 0.028460.30686-0.33212-1.64993 DMTH4 (April) 0.029130.28729-0.04200-0.28663 DMTH5 (May) 0.030180.26660-0.06509-0.38395 DMTH6 (June) 0.001800.020910.273702.17017 DMTH7 (July) -0.03571-0.34487-0.03585-0.27654 DMTH8 (August) 0.078360.675060.289781.79491 DMTH9 (September) 0.084970.804120.321221.77498 DMTH10(October) 0.010470.11194-0.12809-0.96259 DMTH11(November) -0.04511-0.51770-0.04859-0.39734 DMTH12(December) -0.06041-0.701380.002470.01491 PRT -0.01591-6.82772-0.01190-4.27087 MILLS -0.47072-1.14597-1.32969-1.44189 SIGMA 1.4994257.538081.7295542.79992 INDOOR CUT-FLOWERS FREQUENCY NEW ENGLAND

PAGE 194

174 Table A.6 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in New England BETA T-VALUE BETA T-VALUE C -4.05604-0.381310.010800.00600 DINC2 (25/50) -0.46722-1.26049-0.33016-1.52388 DINC3 (50/75) 0.241060.525020.048520.40103 DINC4 (75+) 0.432290.381310.056660.69202 DGEN2 (Female) 0.352320.35132-0.00982-0.03021 DPUR2 (Gifts) 0.332750.13591-0.46145-2.16620 DAGE2 (25/39) 0.660431.20429-0.29294-3.16510 DAGE3 (40/55) 0.278010.165060.195220.39472 DAGE4 (55+) 0.702370.402520.175410.35635 DMTH2 (February) 0.479750.464500.052890.17467 DMTH3 (March) -0.93674-2.08101-0.15526-0.48679 DMTH4 (April) -0.27131-0.350990.139570.78807 DMTH5 (May) 0.667840.605380.013880.05387 DMTH6 (June) -0.48394-0.415170.264361.94404 DMTH7 (July) 0.075220.11450-0.13384-0.92444 DMTH8 (August) 0.988511.317940.130130.69660 DMTH9 (September) 0.325380.540790.206560.79199 DMTH10(October) -0.64799-0.93049-0.08909-0.58898 DMTH11(November) -0.15926-0.245930.050410.31307 DMTH12(December) 0.780881.17970-0.24500-0.81891 PRT -0.00399-0.93826-0.01485-2.93969 MILLS 0.658550.13116-0.73623-0.50170 SIGMA 2.0088913.015061.8515039.44944 ARRANGE NON-ARRAN FREQUENCY NEW ENGLAND

PAGE 195

175 Table A.7 Buyer Frequency Model I Results for Plants and Dry/ Artificial in New England BETA T-VALUE BETA T-VALUE C 1.016460.736939.035040.43971 DINC2 (25/50) -0.03467-0.325170.152910.38392 DINC3 (50/75) 0.141440.874000.216560.36262 DINC4 (75+) -0.05532-0.50014-0.62664-0.47406 DGEN2 (Female) -0.42854-1.18191-1.92797-0.34871 DPUR2 (Gifts) -0.28591-4.214720.704020.48241 DAGE2 (25/39) 0.068550.65925-0.27635-0.27376 DAGE3 (40/55) -0.12733-0.39288-1.14395-0.42789 DAGE4 (55+) -0.25771-0.61985-2.22679-0.60372 DMTH2 (February) -0.09934-0.710770.082840.07971 DMTH3 (March) -0.00151-0.00920-1.08447-0.50914 DMTH4 (April) -0.23477-0.766590.117580.08781 DMTH5 (May) -0.26130-0.82224-0.29199-0.21262 DMTH6 (June) -0.14413-1.02639-0.50522-0.59018 DMTH7 (July) 0.268290.900971.161640.49516 DMTH8 (August) 0.167570.505120.705300.72719 DMTH9 (September) 0.119640.600330.357840.45094 DMTH10(October) 0.280341.38892-0.59686-0.50756 DMTH11(November) -0.11815-0.76052-0.12867-0.11929 DMTH12(December) -0.19870-0.69806-0.16773-0.30902 PRT -0.02485-5.30334-0.06862-4.87350 MILLS -1.45561-1.47720-4.52276-0.50078 SIGMA 1.6322134.766523.0551622.58691 PLANTS DRY FREQUENCY NEW ENGLAND

PAGE 196

176 Table A.8 Buyer Frequency Model I Re sults for Outdoor in New England BETA T-VALUE C -3.39911-4.13783 DINC2 (25/50) 0.127351.55043 DINC3 (50/75) -0.11733-1.44546 DINC4 (75+) 0.114131.33422 DGEN2 (Female) 0.534373.78252 DPUR2 (Gifts) -0.81341-9.35332 DAGE2 (25/39) 0.432813.93537 DAGE3 (40/55) 1.121275.55404 DAGE4 (55+) 1.025774.22271 DMTH2 (February) -1.77181-4.14947 DMTH3 (March) -0.26666-1.32038 DMTH4 (April) 1.588584.74463 DMTH5 (May) 3.002526.20753 DMTH6 (June) 1.989116.41835 DMTH7 (July) 0.913634.77600 DMTH8 (August) 0.341621.88687 DMTH9 (September) 0.564222.78032 DMTH10(October) -0.03051-0.16288 DMTH11(November) -1.36324-2.79655 DMTH12(December) -2.44663-4.93619 PRT -0.02181-5.24252 MILLS 1.697573.41502 SIGMA 1.9336544.95618 OUTDOOR FREQUENCY NEW ENGLAND

PAGE 197

177 Table A.9 Market Penetration Model I Resu lts for Indoor and Cut-Flowers in Middle Atlantic BETA T-VALUE BETA T-VALUE C 0.0012030.408720.000144.82404 DINC2 (25/50) 0.000091.368950.000102.20376 DINC3 (50/75) -0.00090-13.64172-0.00036-7.91661 DINC4 (75+) 0.0011618.039730.0007516.88895 DGEN2 (Female) 0.0018247.779220.0009234.20928 DPUR2 (Gifts) 0.0007820.529990.0011742.49546 DAGE2 (25/39) -0.00003-0.46288-0.00002-0.42574 DAGE3 (40/55) 0.0011518.263950.0009521.48083 DAGE4 (55+) 0.0022936.267500.0013229.96511 DMTH2 (February) 0.000665.452730.0009811.79879 DMTH3 (March) 0.000625.082300.000516.07124 DMTH4 (April) 0.001109.080190.000455.32694 DMTH5 (May) 0.0015913.362130.000809.59226 DMTH6 (June) -0.00047-3.74214-0.00021-2.41705 DMTH7 (July) -0.00087-6.86383-0.00044-5.02023 DMTH8 (August) -0.00085-6.75637-0.00032-3.70099 DMTH9 (September) -0.00052-4.21148-0.00034-3.90730 DMTH10(October) -0.00075-5.97406-0.00036-4.05168 DMTH11(November) -0.00044-3.53854-0.00027-3.04511 DMTH12(December) 0.000756.13720-0.00045-5.01282 SIGMA 0.00331110.979140.0022098.37294 MARKET PENETRATION MIDDLE ATLANTIC INDOOR CUT-FLOWERS

PAGE 198

178 Table A.10 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in Middle Atlantic BETA T-VALUE BETA T-VALUE C -0.00171-36.86998-0.00008-2.68317 DINC2 (25/50) 0.000205.186900.000081.86112 DINC3 (50/75) -0.00012-3.05754-0.00030-7.05995 DINC4 (75+) 0.0004110.643160.0006014.42617 DGEN2 (Female) 0.0004819.744320.0007630.13659 DPUR2 (Gifts) 0.0013740.194320.0008031.12759 DAGE2 (25/39) 0.000225.33599-0.00006-1.40433 DAGE3 (40/55) 0.0006516.524660.0008320.30006 DAGE4 (55+) 0.0006416.168350.0011227.13619 DMTH2 (February) 0.000618.596240.0007910.21205 DMTH3 (March) -0.00004-0.528410.000557.04983 DMTH4 (April) 0.000243.328190.000415.14470 DMTH5 (May) 0.000496.937460.000688.76405 DMTH6 (June) -0.00029-3.57058-0.00014-1.67162 DMTH7 (July) -0.00032-3.98603-0.00036-4.36123 DMTH8 (August) -0.00028-3.43428-0.00024-2.95492 DMTH9 (September) -0.00027-3.38922-0.00026-3.11564 DMTH10(October) -0.00021-2.67211-0.00031-3.73522 DMTH11(November) -0.00002-0.23110-0.00023-2.77345 DMTH12(December) 0.000192.53338-0.00054-6.38504 SIGMA 0.0014761.036160.0020493.14124 MARKET PENETRATION MIDDLE ATLANTIC ARRANGE NON-ARRAN

PAGE 199

179 Table A.11 Market Penetration Model I Results for Plants and Dry/Artificial in Middle Atlantic BETA T-VALUE BETA T-VALUE C -0.00068-19.24644-0.00169-36.45294 DINC2 (25/50) 0.000163.458750.000051.28244 DINC3 (50/75) -0.00056-11.23448-0.00040-9.08517 DINC4 (75+) 0.0005210.858380.000215.16480 DGEN2 (Female) 0.0013344.446510.0012538.43192 DPUR2 (Gifts) -0.00024-8.54856-0.00027-11.03642 DAGE2 (25/39) 0.000102.138880.000020.40166 DAGE3 (40/55) 0.0007816.417760.0004911.79545 DAGE4 (55+) 0.0014430.557230.0010425.39236 DMTH2 (February) -0.00037-3.926620.000293.81946 DMTH3 (March) 0.000353.901770.000192.47933 DMTH4 (April) 0.0012013.874640.000172.23402 DMTH5 (May) 0.0013816.277170.000273.44932 DMTH6 (June) -0.00036-3.74499-0.00032-3.80425 DMTH7 (July) -0.00062-6.31983-0.00047-5.40393 DMTH8 (August) -0.00077-7.76912-0.00030-3.54629 DMTH9 (September) -0.00063-6.447650.000081.07666 DMTH10(October) -0.00071-7.241650.00000-0.05456 DMTH11(November) -0.00044-4.562600.000182.34092 DMTH12(December) 0.0015818.415510.000091.16599 SIGMA 0.0022387.578470.0015659.56383 MARKET PENETRATION MIDDLE ATLANTIC PLANTS DRY

PAGE 200

180 Table A.12 Market Penetration Model I Re sults for Outdoor in Middle Atlantic BETA T-VALUE C -0.00352-39.20637 DINC2 (25/50) 0.000302.93581 DINC3 (50/75) -0.00061-5.74875 DINC4 (75+) 0.0011010.68373 DGEN2 (Female) 0.0021533.45173 DPUR2 (Gifts) -0.00136-22.10889 DAGE2 (25/39) 0.000242.27214 DAGE3 (40/55) 0.0016916.33272 DAGE4 (55+) 0.0027626.92522 DMTH2 (February) -0.00329-12.90445 DMTH3 (March) 0.000402.08927 DMTH4 (April) 0.0038722.30121 DMTH5 (May) 0.0073843.70578 DMTH6 (June) 0.0037821.49651 DMTH7 (July) 0.000532.70998 DMTH8 (August) -0.00086-4.08886 DMTH9 (September) 0.000201.02294 DMTH10(October) -0.00042-2.06566 DMTH11(November) -0.00329-12.84184 DMTH12(December) -0.00340-13.12962 SIGMA 0.0043481.77007 OUTDOOR MARKET PENETRATION MIDDLE ATLANTIC

PAGE 201

181 Table A.13 Buyer Frequency Model I Results for Indoor and Cut-Flowers in Middle Atlantic BETA T-VALUE BETA T-VALUE C 0.163893.24549-0.38860-3.82224 DINC2 (25/50) 0.002060.088390.016050.53818 DINC3 (50/75) -0.04344-1.87760-0.08797-3.12379 DINC4 (75+) 0.034091.432200.157125.70751 DGEN2 (Female) 0.154016.444240.136834.22204 DPUR2 (Gifts) -0.11213-5.126750.076991.67306 DAGE2 (25/39) -0.06140-2.38553-0.07351-2.29016 DAGE3 (40/55) 0.173015.634830.313176.73278 DAGE4 (55+) 0.169935.909400.329347.68826 DMTH2 (February) 0.089612.148850.087071.66442 DMTH3 (March) 0.043981.050440.081561.57876 DMTH4 (April) 0.078241.889200.035300.67500 DMTH5 (May) 0.138593.253840.162123.04419 DMTH6 (June) 0.010780.247760.021380.39439 DMTH7 (July) -0.11123-2.42646-0.07454-1.31672 DMTH8 (August) -0.06770-1.47763-0.03012-0.54441 DMTH9 (September) -0.05799-1.32999-0.01427-0.25905 DMTH10(October) -0.10371-2.27394-0.07912-1.44244 DMTH11(November) -0.02227-0.50459-0.07263-1.31177 DMTH12(December) 0.015730.37511-0.01972-0.34111 PRT -0.00665-5.94688-0.00527-4.70907 MILLS -0.13763-1.348940.266181.87049 SIGMA 0.9484080.294481.0170762.27788 INDOOR CUT-FLOWERS FREQUENCY MIDDLE ATLANTIC

PAGE 202

182 Table A.14 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in Middle Atlantic BETA T-VALUE BETA T-VALUE C -5.35194-3.92432-0.30071-1.65573 DINC2 (25/50) 0.434283.30331-0.04664-1.20694 DINC3 (50/75) -0.13918-1.90380-0.05758-1.75745 DINC4 (75+) 0.424844.391050.167175.25781 DGEN2 (Female) 0.768814.666030.062671.28064 DPUR2 (Gifts) 1.903193.42345-0.04765-0.76372 DAGE2 (25/39) 0.214691.39888-0.14353-3.88057 DAGE3 (40/55) 0.915972.994990.282003.73815 DAGE4 (55+) 0.860274.013030.300885.01299 DMTH2 (February) 0.697303.402940.073781.14783 DMTH3 (March) -0.01346-0.102870.056890.90501 DMTH4 (April) 0.556433.81184-0.00989-0.15776 DMTH5 (May) 0.477142.427790.151902.22035 DMTH6 (June) -0.47731-2.417430.065931.05953 DMTH7 (July) -0.35373-2.10925-0.02544-0.37610 DMTH8 (August) -0.40892-2.47021-0.04341-0.67437 DMTH9 (September) -0.48016-2.626430.021780.34069 DMTH10(October) -0.28652-1.90685-0.02793-0.43846 DMTH11(November) -0.00864-0.06525-0.09678-1.47988 DMTH12(December) 0.452903.41650-0.05665-0.75229 PRT -0.00121-0.81780-0.01045-4.53710 MILLS 2.170662.651930.037510.16386 SIGMA 1.2968925.124351.1161956.90021 ARRANGE NON-ARRAN FREQUENCY MIDDLE ATLANTIC

PAGE 203

183 Table A.15 Buyer Frequency Model I Results for Plants and Dry/Artificial in Middle Atlantic BETA T-VALUE BETA T-VALUE C -0.14382-0.491043.029931.08435 DINC2 (25/50) -0.02000-0.36117-0.31290-1.64876 DINC3 (50/75) -0.07784-1.747240.211761.05921 DINC4 (75+) 0.109472.431720.297132.84096 DGEN2 (Female) 0.076390.72754-0.85931-0.81096 DPUR2 (Gifts) -0.22349-8.764050.020120.10977 DAGE2 (25/39) -0.10746-2.013920.176161.34037 DAGE3 (40/55) 0.127521.34518-0.77199-1.38489 DAGE4 (55+) -0.03398-0.36338-0.64594-0.93257 DMTH2 (February) -0.05390-0.66395-0.08363-0.32806 DMTH3 (March) 0.232203.05982-0.12411-0.64957 DMTH4 (April) 0.306852.94966-0.03008-0.15501 DMTH5 (May) 0.200621.662180.044940.22814 DMTH6 (June) -0.09791-1.123300.676722.23174 DMTH7 (July) -0.08660-0.89679-0.08740-0.23975 DMTH8 (August) -0.20636-1.99570-0.02613-0.10459 DMTH9 (September) -0.07574-0.78489-0.50688-2.50790 DMTH10(October) -0.16904-1.71326-0.11678-0.67879 DMTH11(November) -0.17991-1.978850.150400.77765 DMTH12(December) 0.190531.90434-0.11778-0.64232 PRT -0.01547-6.61332-0.03810-7.90759 MILLS -0.32037-1.03854-2.15219-1.25244 SIGMA 1.2658651.703082.0946440.27804 PLANTS DRY FREQUENCY MIDDLE ATLANTIC

PAGE 204

184 Table A.16 Buyer Frequency Model I Resu lts for Outdoor in Middle Atlantic BETA T-VALUE C -0.93675-3.62507 DINC2 (25/50) 0.045770.91293 DINC3 (50/75) 0.009260.19121 DINC4 (75+) 0.091201.86074 DGEN2 (Female) 0.288064.84300 DPUR2 (Gifts) -0.56934-14.54762 DAGE2 (25/39) 0.250124.38438 DAGE3 (40/55) 0.338354.38381 DAGE4 (55+) 0.376794.74412 DMTH2 (February) -0.89392-4.47784 DMTH3 (March) 0.227692.28315 DMTH4 (April) 1.099717.48672 DMTH5 (May) 1.645278.70096 DMTH6 (June) 1.136648.66836 DMTH7 (July) 0.384773.96624 DMTH8 (August) -0.02342-0.22369 DMTH9 (September) -0.13606-1.34761 DMTH10(October) -0.28343-2.71329 DMTH11(November) -0.66661-3.63877 DMTH12(December) -1.42583-6.02560 PRT -0.02155-7.71120 MILLS 0.595522.73686 SIGMA 1.4541557.51635 OUTDOOR FREQUENCY MIDDLE ATLANTIC

PAGE 205

185 Table A.17 Market Penetration Model I Results for Indoor and Cut-Flowers in East North Central BETA T-VALUE BETA T-VALUE C 0.0010828.849890.000041.71770 DINC2 (25/50) 0.000498.143970.000318.03811 DINC3 (50/75) -0.00086-13.78195-0.00031-8.03863 DINC4 (75+) 0.0007411.970380.0004110.59222 DGEN2 (Female) 0.0018450.984710.0008536.54736 DPUR2 (Gifts) 0.0004913.792160.0008536.21226 DAGE2 (25/39) 0.000183.039570.000277.10488 DAGE3 (40/55) 0.0008514.253280.0006617.47929 DAGE4 (55+) 0.0019733.053400.0007620.07913 DMTH2 (February) 0.000675.861120.0007510.57621 DMTH3 (March) 0.000433.685690.000273.67646 DMTH4 (April) 0.000867.477680.000385.30464 DMTH5 (May) 0.0017415.401070.000689.50339 DMTH6 (June) -0.00037-3.18455-0.00019-2.51701 DMTH7 (July) -0.00079-6.63287-0.00022-2.89587 DMTH8 (August) -0.00080-6.67234-0.00026-3.43505 DMTH9 (September) -0.00073-6.07977-0.00024-3.23063 DMTH10(October) -0.00028-2.400910.000111.53469 DMTH11(November) -0.00045-3.74598-0.00029-3.89062 DMTH12(December) 0.000423.56602-0.00064-8.18635 SIGMA 0.00313109.657510.0018795.00539 MARKET PENETRATION -EAST NORTH CENTRAL INDOOR CUT-FLOWERS

PAGE 206

186 Table A.18 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in East North Central BETA T-VALUE BETA T-VALUE C -0.00131-35.89736-0.00019-7.54994 DINC2 (25/50) 0.000288.786420.000267.23963 DINC3 (50/75) -0.00018-5.21619-0.00026-7.09253 DINC4 (75+) 0.000236.909850.000318.59737 DGEN2 (Female) 0.0004321.095690.0007232.65769 DPUR2 (Gifts) 0.0009938.389470.0005424.63610 DAGE2 (25/39) 0.000185.510830.000226.22446 DAGE3 (40/55) 0.0004212.803790.0006016.86389 DAGE4 (55+) 0.0005215.816050.0005916.52314 DMTH2 (February) 0.000589.961700.000588.69310 DMTH3 (March) -0.00003-0.437410.000304.39420 DMTH4 (April) 0.000243.998850.000324.67090 DMTH5 (May) 0.000549.226890.000527.77248 DMTH6 (June) -0.00019-2.92223-0.00011-1.50818 DMTH7 (July) -0.00033-4.74695-0.00010-1.39934 DMTH8 (August) -0.00028-4.12316-0.00016-2.26369 DMTH9 (September) -0.00023-3.39139-0.00017-2.32341 DMTH10(October) 0.000010.088210.000152.17218 DMTH11(November) -0.00009-1.31592-0.00028-3.92127 DMTH12(December) -0.00006-0.91914-0.00072-9.57155 SIGMA 0.0012760.762230.0017588.68733 MARKET PENETRATION -EAST NORTH CENTRAL ARRANGE NON-ARRAN

PAGE 207

187 Table A.19 Market Penetration Model I Results for Plants and Dry/Artificial in East North Central BETA T-VALUE BETA T-VALUE C -0.00051-15.97114-0.00154-36.24045 DINC2 (25/50) 0.000368.343970.000338.19081 DINC3 (50/75) -0.00051-11.03935-0.00063-13.85691 DINC4 (75+) 0.000337.375490.000122.79280 DGEN2 (Female) 0.0012946.904740.0013643.23946 DPUR2 (Gifts) -0.00019-7.41910-0.00025-10.45042 DAGE2 (25/39) 0.000112.467490.000020.36171 DAGE3 (40/55) 0.0006815.459350.0004610.87622 DAGE4 (55+) 0.0013130.115110.0012530.40286 DMTH2 (February) 0.000111.280030.000101.21290 DMTH3 (March) 0.000252.945320.000283.53798 DMTH4 (April) 0.000809.867240.000202.60740 DMTH5 (May) 0.0013516.965580.000608.01277 DMTH6 (June) -0.00020-2.35786-0.00031-3.72526 DMTH7 (July) -0.00069-7.58321-0.00027-3.21227 DMTH8 (August) -0.00074-8.11541-0.00028-3.37170 DMTH9 (September) -0.00080-8.70746-0.00010-1.23336 DMTH10(October) -0.00062-6.97649-0.00004-0.52665 DMTH11(November) -0.00023-2.644710.000060.76057 DMTH12(December) 0.0012315.35391-0.00002-0.26172 SIGMA 0.0020890.085820.0016566.63228 MARKET PENETRATION -EAST NORTH CENTRAL PLANTS DRY

PAGE 208

188 Table A.20 Market Penetration Model I Resu lts for Outdoor in East North Central BETA T-VALUE C -0.00471-41.64361 DINC2 (25/50) 0.001088.57392 DINC3 (50/75) -0.00107-7.91628 DINC4 (75+) 0.000654.95057 DGEN2 (Female) 0.0027133.77623 DPUR2 (Gifts) -0.00199-25.73450 DAGE2 (25/39) 0.000271.99695 DAGE3 (40/55) 0.0019415.11376 DAGE4 (55+) 0.0034326.96861 DMTH2 (February) -0.00294-10.00443 DMTH3 (March) -0.00075-2.94576 DMTH4 (April) 0.0036116.53444 DMTH5 (May) 0.0094245.29320 DMTH6 (June) 0.0048422.40925 DMTH7 (July) 0.000974.06954 DMTH8 (August) -0.00024-0.95044 DMTH9 (September) 0.000170.68198 DMTH10(October) -0.00081-3.15781 DMTH11(November) -0.00442-13.31975 DMTH12(December) -0.00518-14.54624 SIGMA 0.0053081.71599 MARKET PENETRATION -EAST NORTH CENTRAL OUTDOOR

PAGE 209

189 Table A.21 Buyer Frequency Model I Results for Indoor and Cut-Flow ers in East North Central BETA T-VALUE BETA T-VALUE C 0.202753.563720.047570.37609 DINC2 (25/50) 0.026441.04220-0.00527-0.15081 DINC3 (50/75) -0.09475-3.99306-0.03712-1.19591 DINC4 (75+) 0.038751.502810.178175.87105 DGEN2 (Female) 0.180706.748420.039630.93341 DPUR2 (Gifts) -0.09465-4.49166-0.17016-3.34759 DAGE2 (25/39) 0.054062.06840-0.05563-1.54113 DAGE3 (40/55) 0.191616.328780.098791.96646 DAGE4 (55+) 0.055811.935770.095482.20068 DMTH2 (February) 0.107072.592460.034850.61515 DMTH3 (March) 0.047811.147680.094911.76154 DMTH4 (April) 0.034290.82756-0.03108-0.56121 DMTH5 (May) 0.168644.02926-0.00680-0.12178 DMTH6 (June) -0.04412-1.022620.015100.26654 DMTH7 (July) -0.04072-0.898840.003780.06607 DMTH8 (August) 0.013670.301080.127582.19303 DMTH9 (September) -0.08528-1.854870.066121.12437 DMTH10(October) -0.04951-1.16141-0.09633-1.71726 DMTH11(November) -0.09443-2.13725-0.02010-0.34597 DMTH12(December) -0.03254-0.76635-0.13270-1.90317 PRT -0.00923-7.97627-0.00494-4.01902 MILLS -0.19234-1.73715-0.43971-2.59534 SIGMA 0.9319379.637341.0288358.29476 INDOOR CUT-FLOWERS FREQUENCY EAST NORTH CENTRAL

PAGE 210

190 Table A.22 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in East North Central BETA T-VALUE BETA T-VALUE C -2.69360-1.84538-0.07336-0.28117 DINC2 (25/50) 0.029610.16193-0.01280-0.24215 DINC3 (50/75) -0.09520-0.85949-0.00606-0.14888 DINC4 (75+) 0.323313.800200.175414.72063 DGEN2 (Female) 0.261701.201640.044800.59659 DPUR2 (Gifts) 0.601951.13637-0.18688-2.46409 DAGE2 (25/39) -0.09784-0.69988-0.04078-0.82790 DAGE3 (40/55) 0.365531.310990.105731.21944 DAGE4 (55+) 0.227450.842220.186202.85675 DMTH2 (February) 0.212700.790960.000840.01059 DMTH3 (March) 0.385352.712160.088341.29989 DMTH4 (April) 0.089060.48473-0.03843-0.52599 DMTH5 (May) 0.383321.56457-0.02814-0.35688 DMTH6 (June) -0.21123-1.117600.039750.57654 DMTH7 (July) -0.72090-2.789870.055210.81194 DMTH8 (August) -0.02508-0.122470.116981.65557 DMTH9 (September) 0.073880.405050.126491.70370 DMTH10(October) 0.024040.15566-0.15164-2.14382 DMTH11(November) -0.37538-2.070520.004850.06652 DMTH12(December) 0.216941.42293-0.18063-1.59948 PRT -0.00134-0.77166-0.00837-3.59748 MILLS 0.601650.64274-0.37535-1.22895 SIGMA 1.4782722.671991.1776153.24878 ARRANGE NON-ARRAN FREQUENCY EAST NORTH CENTRAL

PAGE 211

191 Table A.23 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in East North Central BETA T-VALUE BETA T-VALUE C -0.43592-1.800232.532121.86946 DINC2 (25/50) 0.036330.62363-0.37997-1.72570 DINC3 (50/75) -0.16951-3.776170.096460.47675 DINC4 (75+) 0.095712.033820.287722.20071 DGEN2 (Female) 0.167111.75935-0.54003-0.98205 DPUR2 (Gifts) -0.26927-11.666290.086451.16475 DAGE2 (25/39) 0.138602.831140.025280.22598 DAGE3 (40/55) 0.235052.86680-0.31487-1.20083 DAGE4 (55+) 0.107211.23380-0.70621-1.85388 DMTH2 (February) 0.063610.88426-0.05202-0.37629 DMTH3 (March) -0.04913-0.676700.040310.27516 DMTH4 (April) 0.170342.10707-0.02414-0.17470 DMTH5 (May) 0.361743.953540.178880.94814 DMTH6 (June) 0.095391.29654-0.08670-0.49341 DMTH7 (July) -0.12167-1.351640.097970.61631 DMTH8 (August) -0.11221-1.214630.305451.90788 DMTH9 (September) -0.36016-3.746520.142141.02212 DMTH10(October) -0.14808-1.74553-0.00494-0.03597 DMTH11(November) -0.15028-1.94711-0.37652-2.68500 DMTH12(December) 0.271093.35229-0.21788-1.55770 PRT -0.01702-7.86651-0.04855-11.60330 MILLS 0.019630.07429-1.89781-2.05464 SIGMA 1.2123353.210631.8542546.91425 PLANTS DRY FREQUENCY EAST NORTH CENTRAL

PAGE 212

192 Table A.24 Buyer Frequency Model I Results for Outdoor in East North Central BETA T-VALUE C -0.68412-2.81196 DINC2 (25/50) 0.036410.64174 DINC3 (50/75) -0.07017-1.29503 DINC4 (75+) 0.129582.42406 DGEN2 (Female) 0.322265.35748 DPUR2 (Gifts) -0.62174-13.76226 DAGE2 (25/39) 0.268784.77098 DAGE3 (40/55) 0.341134.69397 DAGE4 (55+) 0.291473.53302 DMTH2 (February) -0.88118-5.30719 DMTH3 (March) -0.26206-2.37094 DMTH4 (April) 0.850446.59552 DMTH5 (May) 1.634349.20345 DMTH6 (June) 1.081738.24145 DMTH7 (July) 0.324303.20876 DMTH8 (August) -0.15826-1.51255 DMTH9 (September) -0.01894-0.18363 DMTH10(October) -0.15888-1.44786 DMTH11(November) -0.50656-2.55634 DMTH12(December) -1.03396-4.13014 PRT -0.02638-9.10834 MILLS 0.491842.44378 SIGMA 1.5242658.53256 OUTDOOR FREQUENCY EAST NORTH CENTRAL

PAGE 213

193 Table A.25 Market Penetration Model I Resu lts for Indoor and Cut-Flowers in West North Central BETA T-VALUE BETA T-VALUE C -0.00066-10.10385-0.00191-29.92893 DINC2 (25/50) 0.000828.884360.0008311.53752 DINC3 (50/75) -0.00196-19.68521-0.00113-14.17164 DINC4 (75+) 0.000586.045690.000344.65987 DGEN2 (Female) 0.0025844.680390.0012026.22488 DPUR2 (Gifts) 0.0008515.300480.0014630.95557 DAGE2 (25/39) -0.00012-1.27856-0.00004-0.54224 DAGE3 (40/55) 0.0011512.449070.0009613.28221 DAGE4 (55+) 0.0026529.033140.0011515.93829 DMTH2 (February) 0.000653.697290.001007.46993 DMTH3 (March) 0.000512.890360.000624.53809 DMTH4 (April) 0.001297.437550.000392.82396 DMTH5 (May) 0.0025515.019850.001047.77658 DMTH6 (June) -0.00036-1.94845-0.00017-1.20757 DMTH7 (July) -0.00102-5.50430-0.00040-2.76207 DMTH8 (August) -0.00099-5.30178-0.00025-1.72440 DMTH9 (September) -0.00066-3.58852-0.00027-1.88395 DMTH10(October) -0.00084-4.55553-0.00014-0.95960 DMTH11(November) -0.00057-3.06771-0.00051-3.47914 DMTH12(December) 0.000120.64600-0.00081-5.37538 SIGMA 0.0044589.286020.0031869.83282 MARKET PENETRATION WEST NORTH CENTRAL INDOOR CUT-FLOWERS

PAGE 214

194 Table A.26 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in West North Central BETA T-VALUE BETA T-VALUE C -0.00423-32.53867-0.00236-33.20700 DINC2 (25/50) 0.000728.563370.0007810.77605 DINC3 (50/75) -0.00082-8.51358-0.00108-13.11744 DINC4 (75+) 0.000182.073230.000334.38885 DGEN2 (Female) 0.0007213.498840.0011223.72904 DPUR2 (Gifts) 0.0021026.609740.0009720.68373 DAGE2 (25/39) 0.000131.40114-0.00011-1.45280 DAGE3 (40/55) 0.000799.222560.0008211.16950 DAGE4 (55+) 0.0009811.427380.0010414.20468 DMTH2 (February) 0.001137.596020.000886.50406 DMTH3 (March) -0.00016-0.939410.000775.65593 DMTH4 (April) 0.000020.119470.000483.47592 DMTH5 (May) 0.000724.638030.001007.42310 DMTH6 (June) -0.00004-0.22001-0.00015-1.03941 DMTH7 (July) -0.00067-3.66570-0.00023-1.59005 DMTH8 (August) -0.00050-2.79888-0.00017-1.17638 DMTH9 (September) -0.00038-2.15737-0.00022-1.52466 DMTH10(October) 0.000110.69457-0.00015-1.00878 DMTH11(November) -0.00009-0.51329-0.00060-3.91732 DMTH12(December) 0.000000.00385-0.00107-6.65342 SIGMA 0.0028143.682320.0031161.75167 MARKET PENETRATION WEST NORTH CENTRAL ARRANGE NON-ARRAN

PAGE 215

195 Table A.27 Market Penetration Model I Results for Plants and Dry/Artificial in West North Central BETA T-VALUE BETA T-VALUE C -0.00275-34.43173-0.00435-35.81251 DINC2 (25/50) 0.000516.115850.000495.52885 DINC3 (50/75) -0.00127-13.55338-0.00173-16.06732 DINC4 (75+) 0.000293.311660.000434.75988 DGEN2 (Female) 0.0020636.104900.0026533.89896 DPUR2 (Gifts) -0.00030-6.04830-0.00034-6.54131 DAGE2 (25/39) -0.00007-0.841710.000010.11376 DAGE3 (40/55) 0.0010312.210650.000818.72858 DAGE4 (55+) 0.0019923.806380.0024927.26559 DMTH2 (February) -0.00003-0.19210-0.00005-0.30628 DMTH3 (March) 0.000150.923970.000543.27954 DMTH4 (April) 0.0015510.260570.000432.57167 DMTH5 (May) 0.0021314.365820.0015710.18304 DMTH6 (June) -0.00017-1.02031-0.00011-0.61164 DMTH7 (July) -0.00081-4.64900-0.00089-4.74168 DMTH8 (August) -0.00110-6.11824-0.00086-4.52883 DMTH9 (September) -0.00079-4.521450.000060.32366 DMTH10(October) -0.00113-6.27106-0.00063-3.45443 DMTH11(November) -0.00045-2.655150.000291.72278 DMTH12(December) 0.001278.24556-0.00012-0.68351 SIGMA 0.0035366.298080.0030552.04339 MARKET PENETRATION WEST NORTH CENTRAL PLANTS DRY

PAGE 216

196 Table A.28 Market Penetration Model I Resu lts for Outdoor in West North Central BETA T-VALUE C -0.00988-42.93800 DINC2 (25/50) 0.001025.05824 DINC3 (50/75) -0.00220-9.90990 DINC4 (75+) 0.000582.75108 DGEN2 (Female) 0.0039929.83480 DPUR2 (Gifts) -0.00307-24.19862 DAGE2 (25/39) 0.000331.54887 DAGE3 (40/55) 0.0026712.90781 DAGE4 (55+) 0.0046522.84548 DMTH2 (February) -0.00349-7.18557 DMTH3 (March) -0.00006-0.15769 DMTH4 (April) 0.0069020.51203 DMTH5 (May) 0.0131040.42034 DMTH6 (June) 0.0067419.97368 DMTH7 (July) 0.000972.47164 DMTH8 (August) -0.00157-3.59007 DMTH9 (September) 0.000501.26462 DMTH10(October) -0.00129-2.99260 DMTH11(November) -0.00558-9.97931 DMTH12(December) -0.00865-11.97842 SIGMA 0.0073464.43743 MARKET PENETRATION WEST NORTH CENTRAL OUTDOOR

PAGE 217

197 Table A.29 Buyer Frequency Model I Results for Indoor and Cut-Flow ers in West North Central BETA T-VALUE BETA T-VALUE C -0.90320-4.21772-1.47287-2.17034 DINC2 (25/50) 0.304675.218270.316102.25166 DINC3 (50/75) -0.35821-5.80387-0.22209-1.65087 DINC4 (75+) -0.09416-1.816550.039170.65484 DGEN2 (Female) 0.509616.245460.325632.22634 DPUR2 (Gifts) -0.06946-1.702860.038300.21959 DAGE2 (25/39) 0.082271.97094-0.07079-1.21329 DAGE3 (40/55) 0.323384.866850.283201.95848 DAGE4 (55+) 0.347565.173980.284181.96043 DMTH2 (February) 0.224333.264340.195161.43806 DMTH3 (March) 0.115791.735660.164671.49678 DMTH4 (April) 0.240363.450540.180451.72988 DMTH5 (May) 0.462676.160770.303152.53963 DMTH6 (June) -0.00525-0.07504-0.10126-0.97764 DMTH7 (July) -0.10441-1.379420.077550.69597 DMTH8 (August) -0.21042-2.731630.004630.04440 DMTH9 (September) -0.05880-0.80935-0.36429-3.11933 DMTH10(October) -0.17812-2.44310-0.03852-0.37947 DMTH11(November) -0.19665-2.68730-0.23036-1.86860 DMTH12(December) -0.18305-2.57578-0.06963-0.49806 PRT -0.00718-4.09064-0.00040-0.21485 MILLS 0.642992.650900.504430.90793 SIGMA 1.1978058.059141.3144436.39877 INDOOR CUT-FLOWERS FREQUENCY WEST NORTH CENTRAL

PAGE 218

198 Table A.30 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in West North Central BETA T-VALUE BETA T-VALUE C -9.79535-2.209010.857670.58326 DINC2 (25/50) 0.989001.91386-0.14569-0.55143 DINC3 (50/75) -0.91375-1.880220.253430.91729 DINC4 (75+) 0.110980.934860.097551.32256 DGEN2 (Female) 1.105652.32583-0.19072-0.65003 DPUR2 (Gifts) 2.275751.69396-0.45568-1.79363 DAGE2 (25/39) -0.15919-0.950670.022350.28157 DAGE3 (40/55) 1.114861.91534-0.08785-0.35035 DAGE4 (55+) 1.476042.18511-0.26437-0.96751 DMTH2 (February) 1.759202.34547-0.30559-1.22807 DMTH3 (March) -0.09558-0.39174-0.06736-0.33241 DMTH4 (April) 0.174080.797240.041790.23383 DMTH5 (May) 1.127822.81377-0.08429-0.38116 DMTH6 (June) -0.28712-1.17312-0.05229-0.38397 DMTH7 (July) -0.67827-1.409670.278251.92576 DMTH8 (August) -0.26497-0.711320.068780.50515 DMTH9 (September) -1.15393-3.00356-0.24006-1.53620 DMTH10(October) 0.137600.56615-0.04205-0.31017 DMTH11(November) -0.37022-1.44610-0.02469-0.12154 DMTH12(December) 0.485002.370600.300220.95824 PRT 0.003421.52212-0.01803-3.55388 MILLS 4.000171.76865-1.33335-1.22100 SIGMA 1.5710716.175621.5028030.72303 ARRANGE NON-ARRAN FREQUENCY WEST NORTH CENTRAL

PAGE 219

199 Table A.31 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in West North Central BETA T-VALUE BETA T-VALUE C -0.43239-0.42660-4.97111-1.98524 DINC2 (25/50) 0.017640.124040.621672.30634 DINC3 (50/75) -0.16543-0.91271-1.15486-2.19529 DINC4 (75+) 0.158211.693740.072500.65813 DGEN2 (Female) -0.05183-0.169302.039162.27299 DPUR2 (Gifts) -0.37575-8.34657-0.26906-2.41793 DAGE2 (25/39) -0.06187-0.777020.406533.03069 DAGE3 (40/55) -0.06668-0.305620.897732.31875 DAGE4 (55+) 0.021290.083771.276201.78340 DMTH2 (February) 0.002040.016560.168950.92283 DMTH3 (March) -0.10136-0.783550.453102.01003 DMTH4 (April) 0.046740.217940.624983.06524 DMTH5 (May) 0.362661.417841.116323.29135 DMTH6 (June) 0.178771.47610-0.15639-0.82627 DMTH7 (July) -0.06180-0.38765-0.55035-1.82376 DMTH8 (August) -0.27298-1.31819-0.55251-1.62798 DMTH9 (September) 0.030410.19291-0.12467-0.65953 DMTH10(October) 0.045670.23179-0.61264-2.54391 DMTH11(November) -0.29243-2.091970.192070.94444 DMTH12(December) 0.067080.42460-0.33185-1.72672 PRT -0.00957-2.46394-0.05733-8.16575 MILLS -0.38518-0.517272.443361.85709 SIGMA 1.4523734.204642.0115535.67820 PLANTS DRY FREQUENCY WEST NORTH CENTRAL

PAGE 220

200 Table A.32 Buyer Frequency Model I Results for Outdoor in West North Central BETA T-VALUE C -1.79203-3.08206 DINC2 (25/50) 0.085350.95970 DINC3 (50/75) -0.02390-0.24052 DINC4 (75+) -0.06076-0.70455 DGEN2 (Female) 0.575644.77988 DPUR2 (Gifts) -0.85513-9.56936 DAGE2 (25/39) 0.001760.01984 DAGE3 (40/55) 0.292012.36536 DAGE4 (55+) 0.447123.12228 DMTH2 (February) -0.67996-2.50882 DMTH3 (March) 0.053810.31262 DMTH4 (April) 1.535745.81290 DMTH5 (May) 2.179425.90819 DMTH6 (June) 1.371325.36986 DMTH7 (July) 0.053370.31340 DMTH8 (August) -0.28502-1.50670 DMTH9 (September) 0.005620.03285 DMTH10(October) -0.33747-1.83300 DMTH11(November) -1.17148-3.28820 DMTH12(December) -1.99033-3.26207 PRT -0.01441-3.68458 MILLS 0.915012.56206 SIGMA 1.8567447.68707 OUTDOOR FREQUENCY WEST NORTH CENTRAL

PAGE 221

201 Table A.33 Market Penetration Model I Resu lts for Indoor and Cut-Flowers in South Atlantic BETA T-VALUE BETA T-VALUE C 0.0010929.88983-0.00003-1.17538 DINC2 (25/50) 0.000539.138400.0003810.31657 DINC3 (50/75) -0.00122-20.10271-0.00061-15.71302 DINC4 (75+) 0.0009015.160850.0006116.35724 DGEN2 (Female) 0.0017549.919970.0006930.61863 DPUR2 (Gifts) 0.0004212.125340.0008436.63333 DAGE2 (25/39) 0.000152.508980.000277.22679 DAGE3 (40/55) 0.0007212.418270.0005013.51724 DAGE4 (55+) 0.0021036.266980.0009024.48542 DMTH2 (February) 0.000524.683850.000679.65168 DMTH3 (March) 0.000121.055880.000050.72825 DMTH4 (April) 0.000958.551000.000344.87786 DMTH5 (May) 0.001079.650710.000618.72815 DMTH6 (June) -0.00047-4.06703-0.00018-2.46683 DMTH7 (July) -0.00066-5.70532-0.00029-4.01054 DMTH8 (August) -0.00074-6.37325-0.00022-3.05619 DMTH9 (September) -0.00060-5.19680-0.00019-2.56537 DMTH10(October) -0.00034-2.97730-0.00016-2.15810 DMTH11(November) -0.00023-1.96402-0.00013-1.74504 DMTH12(December) 0.001089.70393-0.00028-3.81903 SIGMA 0.00304109.930230.0018193.56892 MARKET PENETRATION SOUTH ATLANTIC INDOOR CUT-FLOWERS

PAGE 222

202 Table A.34 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in South Atlantic BETA T-VALUE BETA T-VALUE C -0.00148-36.08170-0.00030-11.96086 DINC2 (25/50) 0.000329.122050.000319.04224 DINC3 (50/75) -0.00028-7.39752-0.00054-14.57418 DINC4 (75+) 0.000267.277190.0005415.54380 DGEN2 (Female) 0.0004218.993090.0005425.81310 DPUR2 (Gifts) 0.0011839.304010.0004621.60681 DAGE2 (25/39) 0.000215.621680.000236.64390 DAGE3 (40/55) 0.0004211.726340.0004111.96671 DAGE4 (55+) 0.0005515.265720.0007521.72730 DMTH2 (February) 0.000538.138820.000538.22976 DMTH3 (March) -0.00010-1.393860.000121.80549 DMTH4 (April) 0.000263.877350.000284.23816 DMTH5 (May) 0.000507.656550.000446.72607 DMTH6 (June) -0.00018-2.40009-0.00012-1.74899 DMTH7 (July) -0.00023-3.18181-0.00023-3.38491 DMTH8 (August) -0.00018-2.44005-0.00018-2.55908 DMTH9 (September) -0.00018-2.48672-0.00015-2.23532 DMTH10(October) -0.00025-3.35088-0.00008-1.25185 DMTH11(November) -0.00004-0.52334-0.00009-1.37832 DMTH12(December) 0.000060.86145-0.00035-4.94535 SIGMA 0.0013960.578370.0016785.81778 MARKET PENETRATION SOUTH ATLANTIC ARRANGE NON-ARRAN

PAGE 223

203 Table A.35 Market Penetration Model I Results for Plants and Dry/Artificial in South Atlantic BETA T-VALUE BETA T-VALUE C -0.00055-16.44369-0.00133-34.52126 DINC2 (25/50) 0.000449.611330.000184.76446 DINC3 (50/75) -0.00075-15.55526-0.00065-15.23519 DINC4 (75+) 0.000449.393420.000123.10856 DGEN2 (Female) 0.0013547.138770.0013444.83553 DPUR2 (Gifts) -0.00033-12.17884-0.00018-8.01604 DAGE2 (25/39) 0.000071.506360.000010.13808 DAGE3 (40/55) 0.0006514.135280.000348.84865 DAGE4 (55+) 0.0015634.513030.0009725.41038 DMTH2 (February) -0.00005-0.512910.000091.23263 DMTH3 (March) 0.000070.815430.000152.04833 DMTH4 (April) 0.0009611.374630.000162.11915 DMTH5 (May) 0.0008910.622420.000202.67953 DMTH6 (June) -0.00028-3.13437-0.00027-3.50580 DMTH7 (July) -0.00053-5.75174-0.00022-2.88495 DMTH8 (August) -0.00083-8.63777-0.00019-2.47508 DMTH9 (September) -0.00063-6.70731-0.00018-2.26553 DMTH10(October) -0.00033-3.617360.000081.06799 DMTH11(November) -0.00015-1.628700.000070.96501 DMTH12(December) 0.0015318.529700.000435.98317 SIGMA 0.0021589.858780.0015767.90499 MARKET PENETRATION SOUTH ATLANTIC PLANTS DRY

PAGE 224

204 Table A.36 Market Penetration Model I Re sults for Outdoor in South Atlantic BETA T-VALUE C -0.00170-29.07345 DINC2 (25/50) 0.000689.05038 DINC3 (50/75) -0.00105-13.06651 DINC4 (75+) 0.000648.32073 DGEN2 (Female) 0.0017537.62597 DPUR2 (Gifts) -0.00169-36.52163 DAGE2 (25/39) 0.000020.31362 DAGE3 (40/55) 0.0012115.83538 DAGE4 (55+) 0.0025533.91842 DMTH2 (February) -0.00061-4.00808 DMTH3 (March) 0.001178.40597 DMTH4 (April) 0.0034225.91430 DMTH5 (May) 0.0037228.42745 DMTH6 (June) 0.000896.28795 DMTH7 (July) -0.00068-4.40271 DMTH8 (August) -0.00160-9.80351 DMTH9 (September) -0.00033-2.23760 DMTH10(October) 0.000010.04601 DMTH11(November) -0.00125-7.81874 DMTH12(December) -0.00215-12.67922 SIGMA 0.0034489.69334 MARKET PENETRATION SOUTH ATLANTIC OUTDOOR

PAGE 225

205 Table A.37 Buyer Frequency Model I Results for Indoor and Cut-Flowers in South Atlantic BETA T-VALUE BETA T-VALUE C 0.223583.495000.617743.43675 DINC2 (25/50) -0.02716-0.94277-0.17729-3.22234 DINC3 (50/75) -0.13735-5.13940-0.08002-1.70384 DINC4 (75+) 0.132284.893700.310759.03114 DGEN2 (Female) 0.161125.35733-0.15014-2.78943 DPUR2 (Gifts) -0.14500-6.88140-0.38396-5.97647 DAGE2 (25/39) -0.01284-0.47499-0.16471-3.81395 DAGE3 (40/55) 0.109603.27375-0.14847-2.43420 DAGE4 (55+) 0.152184.182860.156462.29144 DMTH2 (February) 0.039850.878040.011540.17348 DMTH3 (March) 0.063981.411270.020280.30852 DMTH4 (April) 0.033670.74390-0.13792-2.06317 DMTH5 (May) 0.122722.74541-0.08837-1.30064 DMTH6 (June) -0.01520-0.325000.090461.34413 DMTH7 (July) -0.04530-0.941180.063540.91389 DMTH8 (August) 0.009750.200580.025010.35699 DMTH9 (September) -0.06315-1.29346-0.00917-0.13364 DMTH10(October) -0.09066-1.94848-0.15103-2.20344 DMTH11(November) -0.00954-0.20370-0.07865-1.15841 DMTH12(December) 0.065251.454260.237663.34387 PRT -0.01067-8.82503-0.00955-6.89506 MILLS -0.11121-0.89941-1.15491-4.98241 SIGMA 1.0123479.602221.2022456.87569 CUT-FLOWERS INDOOR FREQUENCY SOUTH ATLANTIC

PAGE 226

206 Table A.38 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in South Atlantic BETA T-VALUE BETA T-VALUE C -2.15868-1.337941.383453.16081 DINC2 (25/50) 0.158980.75562-0.38528-3.55203 DINC3 (50/75) -0.08386-0.683060.022280.25534 DINC4 (75+) 0.252343.090910.247255.32114 DGEN2 (Female) 0.260241.25634-0.34432-3.30549 DPUR2 (Gifts) 0.368330.59175-0.55387-5.60777 DAGE2 (25/39) 0.039530.28192-0.26716-4.01970 DAGE3 (40/55) 0.084050.29941-0.29760-2.60327 DAGE4 (55+) 0.367301.31463-0.02775-0.20369 DMTH2 (February) 0.194680.78267-0.19065-1.89878 DMTH3 (March) 0.067450.46171-0.04122-0.49650 DMTH4 (April) 0.113000.67920-0.19646-2.18722 DMTH5 (May) 0.303751.46938-0.28421-2.76713 DMTH6 (June) 0.042660.253580.096191.15225 DMTH7 (July) -0.08837-0.492650.190282.12152 DMTH8 (August) -0.14607-0.851640.139641.53762 DMTH9 (September) -0.11384-0.625070.002780.03180 DMTH10(October) -0.51086-2.65200-0.09002-1.06449 DMTH11(November) -0.04278-0.29930-0.05103-0.58828 DMTH12(December) 0.386292.903250.366943.32943 PRT 0.002141.49269-0.01653-5.81936 MILLS 0.189260.18491-1.97103-4.05950 SIGMA 1.3601623.008781.3943351.33488 ARRANGE NON-ARRAN FREQUENCY SOUTH ATLANTIC

PAGE 227

207 Table A.39 Buyer Frequency Model I Results for Plants and Dry/Artificial in South Atlantic BETA T-VALUE BETA T-VALUE C -1.00901-4.059610.256720.18960 DINC2 (25/50) 0.091061.50025-0.08804-0.51515 DINC3 (50/75) -0.08593-1.80291-0.14248-0.68497 DINC4 (75+) 0.044540.976800.175951.46793 DGEN2 (Female) 0.364153.833760.302900.50117 DPUR2 (Gifts) -0.40335-15.55292-0.12678-1.90060 DAGE2 (25/39) 0.039170.820990.051740.60612 DAGE3 (40/55) 0.251133.059930.007300.02934 DAGE4 (55+) 0.333723.388820.013930.04063 DMTH2 (February) -0.08254-1.09252-0.10594-0.76691 DMTH3 (March) 0.089341.238640.379282.72980 DMTH4 (April) 0.297923.661020.156531.08436 DMTH5 (May) 0.399624.93364-0.04954-0.34922 DMTH6 (June) -0.07923-1.03923-0.14131-0.82197 DMTH7 (July) -0.21098-2.42418-0.22526-1.37870 DMTH8 (August) -0.16945-1.743370.038850.24832 DMTH9 (September) -0.12601-1.465320.142470.86078 DMTH10(October) -0.11657-1.51323-0.01002-0.06521 DMTH11(November) -0.06904-0.918580.006360.04625 DMTH12(December) 0.409074.71699-0.12028-0.76362 PRT -0.01359-6.00603-0.05882-12.72530 MILLS 0.646272.42642-0.16946-0.18034 SIGMA 1.2322752.860491.8937047.15032 DRY PLANTS FREQUENCY SOUTH ATLANTIC

PAGE 228

208 Table A.40 Buyer Frequency Model I Resu lts for Outdoor in South Atlantic BETA T-VALUE C -0.51356-2.49747 DINC2 (25/50) 0.163423.06424 DINC3 (50/75) -0.15850-3.19362 DINC4 (75+) 0.074551.56623 DGEN2 (Female) 0.277105.08498 DPUR2 (Gifts) -0.64760-12.20366 DAGE2 (25/39) 0.163783.30484 DAGE3 (40/55) 0.408735.81932 DAGE4 (55+) 0.247452.82457 DMTH2 (February) -0.06788-0.75209 DMTH3 (March) 0.428344.87704 DMTH4 (April) 0.821427.35119 DMTH5 (May) 0.863687.23561 DMTH6 (June) 0.332553.94929 DMTH7 (July) -0.20627-2.20656 DMTH8 (August) -0.13456-1.23917 DMTH9 (September) -0.26045-2.93483 DMTH10(October) -0.09571-1.10820 DMTH11(November) -0.31451-3.06341 DMTH12(December) -0.83150-6.09717 PRT -0.01449-7.05338 MILLS 0.583632.79795 SIGMA 1.5107665.13979 OUTDOOR FREQUENCY SOUTH ATLANTIC

PAGE 229

209 Table A.41 Market Penetration Model I Results for Indoor and Cut-Flowers in East South Central BETA T-VALUE BETA T-VALUE C -0.00135-17.70948-0.00363-34.73145 DINC2 (25/50) 0.000868.442970.000474.61196 DINC3 (50/75) -0.00141-13.08133-0.00030-2.87793 DINC4 (75+) -0.00032-2.95660-0.00035-3.28152 DGEN2 (Female) 0.0024638.600190.0009114.62632 DPUR2 (Gifts) 0.0008413.705660.0017325.77127 DAGE2 (25/39) 0.000595.779030.000515.02069 DAGE3 (40/55) 0.001009.786040.000908.91408 DAGE4 (55+) 0.0023923.678840.000949.25411 DMTH2 (February) 0.001075.581380.001799.94990 DMTH3 (March) 0.000241.205350.000080.38878 DMTH4 (April) 0.001387.196490.000412.15178 DMTH5 (May) 0.001879.861770.001297.03924 DMTH6 (June) -0.00050-2.49690-0.00060-2.88258 DMTH7 (July) -0.00045-2.214850.000030.13703 DMTH8 (August) -0.00100-4.87938-0.00030-1.46909 DMTH9 (September) -0.00114-5.51011-0.00054-2.62399 DMTH10(October) -0.00077-3.80368-0.00015-0.75751 DMTH11(November) -0.00070-3.44754-0.00085-4.02786 DMTH12(December) 0.000844.28831-0.00079-3.73223 SIGMA 0.0048182.726270.0041057.09060 MARKET PENETRATION EAST SOUTH CENTRAL INDOOR CUT-FLOWERS

PAGE 230

210 Table A.42 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in East South Central BETA T-VALUE BETA T-VALUE C -0.00652-29.67010-0.00438-33.76512 DINC2 (25/50) 0.000755.981500.000252.34046 DINC3 (50/75) -0.00036-2.68881-0.00026-2.30479 DINC4 (75+) -0.00014-1.02059-0.00036-3.15161 DGEN2 (Female) 0.0009011.143460.0007110.69065 DPUR2 (Gifts) 0.0027522.553600.0010014.74089 DAGE2 (25/39) 0.000614.585730.000393.55966 DAGE3 (40/55) 0.000826.231650.000817.58581 DAGE4 (55+) 0.0013710.533420.000494.50418 DMTH2 (February) 0.001757.999950.001668.78943 DMTH3 (March) -0.00039-1.468860.000341.64550 DMTH4 (April) 0.000522.162720.000301.44947 DMTH5 (May) 0.001064.636960.001326.87618 DMTH6 (June) -0.00126-4.32277-0.00028-1.26913 DMTH7 (July) 0.000271.10753-0.00011-0.51880 DMTH8 (August) -0.00050-1.86089-0.00020-0.94079 DMTH9 (September) -0.00080-2.88516-0.00032-1.45188 DMTH10(October) 0.000080.31711-0.00016-0.76489 DMTH11(November) -0.00070-2.57596-0.00074-3.25861 DMTH12(December) -0.00009-0.33823-0.00117-4.87906 SIGMA 0.0039337.721240.0040946.68289 ARRANGE NON-ARRAN MARKET PENETRATION EAST SOUTH CENTRAL

PAGE 231

211 Table A.43 Market Penetration Model I Results for Plants and Dry/Artificial in East South Central BETA T-VALUE BETA T-VALUE C -0.00362-34.75237-0.00456-34.68028 DINC2 (25/50) 0.000888.802570.000747.51828 DINC3 (50/75) -0.00141-12.52207-0.00162-13.66532 DINC4 (75+) -0.00016-1.45431-0.00004-0.42400 DGEN2 (Female) 0.0022231.985040.0025331.70697 DPUR2 (Gifts) -0.00038-6.273070.000162.72402 DAGE2 (25/39) 0.000625.986660.000373.50025 DAGE3 (40/55) 0.000676.535250.000888.46712 DAGE4 (55+) 0.0018618.532600.0023222.57570 DMTH2 (February) -0.00010-0.507510.000231.21841 DMTH3 (March) 0.000160.826790.000452.40977 DMTH4 (April) 0.001679.199180.000371.95025 DMTH5 (May) 0.001598.792680.001136.26084 DMTH6 (June) -0.00012-0.61133-0.00029-1.46258 DMTH7 (July) -0.00047-2.28347-0.00030-1.48346 DMTH8 (August) -0.00122-5.61825-0.00077-3.66987 DMTH9 (September) -0.00112-5.22911-0.00047-2.29787 DMTH10(October) -0.00105-4.91100-0.00038-1.89948 DMTH11(November) -0.00037-1.83503-0.00009-0.47661 DMTH12(December) 0.0018710.363230.000452.40378 SIGMA 0.0041059.691740.0035850.56742 MARKET PENETRATION EAST SOUTH CENTRAL PLANTS DRY

PAGE 232

212 Table A.44 Market Penetration Model I Resu lts for Outdoor in East South Central BETA T-VALUE C -0.00798-40.84736 DINC2 (25/50) 0.001106.41050 DINC3 (50/75) -0.00246-12.56430 DINC4 (75+) 0.000050.29628 DGEN2 (Female) 0.0033129.07007 DPUR2 (Gifts) -0.00304-27.33782 DAGE2 (25/39) 0.000351.92993 DAGE3 (40/55) 0.0019611.11885 DAGE4 (55+) 0.0039522.80589 DMTH2 (February) -0.00234-6.06090 DMTH3 (March) 0.001464.54999 DMTH4 (April) 0.0071825.48381 DMTH5 (May) 0.0081229.09273 DMTH6 (June) 0.002377.56850 DMTH7 (July) -0.00120-3.31874 DMTH8 (August) -0.00282-7.05198 DMTH9 (September) 0.000260.76308 DMTH10(October) 0.000621.86028 DMTH11(November) -0.00269-6.76230 DMTH12(December) -0.00629-12.01677 SIGMA 0.0063661.08002 MARKET PENETRATION EAST SOUTH CENTRAL OUTDOOR

PAGE 233

213 Table A.45 Buyer Frequency Model I Results for Indoor and Cut-Flow ers in East South Central BETA T-VALUE BETA T-VALUE C -2.45694-5.00806-11.21046-3.84988 DINC2 (25/50) 0.325323.272040.729132.70799 DINC3 (50/75) -0.26993-3.503220.163631.51427 DINC4 (75+) -0.36803-2.89391-1.09779-3.65673 DGEN2 (Female) 0.984606.071241.243173.62053 DPUR2 (Gifts) 0.065050.915442.206013.28721 DAGE2 (25/39) 0.301543.748260.513212.08530 DAGE3 (40/55) 0.654285.231501.421973.27647 DAGE4 (55+) 0.650634.904701.155183.33485 DMTH2 (February) 0.490294.329521.958873.54179 DMTH3 (March) 0.223142.18278-0.05538-0.29419 DMTH4 (April) 0.355843.108410.578102.46614 DMTH5 (May) 0.773076.088071.797123.95164 DMTH6 (June) -0.42096-3.71839-0.99832-3.40122 DMTH7 (July) -0.05582-0.523340.003400.01841 DMTH8 (August) -0.50448-3.96494-0.36862-1.72509 DMTH9 (September) -0.11224-0.91735-0.35989-1.47618 DMTH10(October) -0.15054-1.331930.171080.95763 DMTH11(November) -0.23490-2.06037-1.32119-3.67530 DMTH12(December) -0.00066-0.00631-1.03809-2.69354 PRT -0.01408-6.58924-0.00884-2.94314 MILLS 2.169974.419806.977823.48210 SIGMA 1.7313554.138351.9266727.38020 INDOOR CUT-FLOWERS FREQUENCY EAST SOUTH CENTRAL

PAGE 234

214 Table A.46 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in East South Central BETA T-VALUE BETA T-VALUE C -33.36719-3.51000-21.55066-2.49494 DINC2 (25/50) 2.585713.109370.893241.94426 DINC3 (50/75) -0.67016-2.257460.082090.40836 DINC4 (75+) -0.80803-2.27564-1.88229-2.78715 DGEN2 (Female) 2.907913.408561.921722.48683 DPUR2 (Gifts) 8.507073.220852.462692.20190 DAGE2 (25/39) 2.415283.260250.627761.32164 DAGE3 (40/55) 3.071273.201322.272522.20786 DAGE4 (55+) 4.213943.183661.520052.80323 DMTH2 (February) 4.900713.236843.473712.24733 DMTH3 (March) -0.79233-1.847230.245400.61175 DMTH4 (April) 0.620381.225391.235052.61660 DMTH5 (May) 3.632013.780393.480722.54303 DMTH6 (June) -3.36574-2.98068-1.09811-2.55782 DMTH7 (July) 0.735441.62029-0.21737-0.79141 DMTH8 (August) -1.54677-2.64862-0.32403-1.03683 DMTH9 (September) -2.30283-3.03184-0.25923-0.76628 DMTH10(October) 0.322141.008060.161880.67527 DMTH11(November) -2.39824-3.30225-1.87094-2.37021 DMTH12(December) -0.14562-0.42811-3.25328-2.38997 PRT -0.00149-0.39774-0.00820-1.37146 MILLS 14.927603.3358312.633332.33383 SIGMA 1.8830513.349862.1574822.91267 ARRANGE NON-ARRAN FREQUENCY EAST SOUTH CENTRAL

PAGE 235

215 Table A.47 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in East South Central BETA T-VALUE BETA T-VALUE C -5.89800-3.09828-9.01252-2.07811 DINC2 (25/50) 0.484461.825340.904341.89341 DINC3 (50/75) -0.71486-2.65837-1.60073-2.08290 DINC4 (75+) -0.15560-0.77912-0.14026-0.51217 DGEN2 (Female) 1.489412.780723.093292.30623 DPUR2 (Gifts) -0.60522-7.63707-0.00827-0.07334 DAGE2 (25/39) 0.457552.592450.878412.86085 DAGE3 (40/55) 0.760212.971321.403362.32607 DAGE4 (55+) 0.949372.442412.140391.95048 DMTH2 (February) 0.319902.03438-0.02279-0.08587 DMTH3 (March) 0.181131.081031.141043.78050 DMTH4 (April) 0.903602.821430.626222.35950 DMTH5 (May) 1.139573.234571.941904.28394 DMTH6 (June) -0.03351-0.20670-1.11942-3.98568 DMTH7 (July) -0.13249-0.72305-0.07855-0.29515 DMTH8 (August) -1.26382-3.75272-0.96555-2.05247 DMTH9 (September) -0.74761-2.69011-0.21167-0.61859 DMTH10(October) -0.56684-2.06973-0.26666-0.90369 DMTH11(November) -0.15283-0.85243-0.16414-0.64439 DMTH12(December) 0.896252.44029-0.26987-0.94466 PRT -0.00812-2.34141-0.05496-8.76763 MILLS 3.277442.522394.843642.00190 SIGMA 1.7915332.551922.6635634.78875 PLANTS DRY FREQUENCY EAST SOUTH CENTRAL

PAGE 236

216 Table A.48 Buyer Frequency Model I Results for Outdoor in East South Central BETA T-VALUE C -3.97547-5.19925 DINC2 (25/50) 0.203021.87496 DINC3 (50/75) -0.66017-4.90933 DINC4 (75+) 0.011010.10456 DGEN2 (Female) 0.964756.02429 DPUR2 (Gifts) -1.25612-8.72988 DAGE2 (25/39) 0.582916.24368 DAGE3 (40/55) 0.823365.56945 DAGE4 (55+) 0.602493.45465 DMTH2 (February) -0.26351-1.22569 DMTH3 (March) 0.692943.95028 DMTH4 (April) 2.291657.07007 DMTH5 (May) 2.406786.69894 DMTH6 (June) 1.161716.43178 DMTH7 (July) -0.00024-0.00134 DMTH8 (August) -0.94626-3.90026 DMTH9 (September) 0.326932.00759 DMTH10(October) 0.432402.66507 DMTH11(November) -1.50989-5.76366 DMTH12(December) -2.72530-5.45768 PRT -0.02166-5.32989 MILLS 2.203504.62410 SIGMA 1.8823644.28353 OUTDOOR FREQUENCY EAST SOUTH CENTRAL

PAGE 237

217 Table A.49 Market Penetration Model I Resu lts for Indoor and Cut-Flowers in West South Central BETA T-VALUE BETA T-VALUE C 0.000265.95169-0.00101-23.88579 DINC2 (25/50) 0.0007711.630710.0005610.86440 DINC3 (50/75) -0.00139-19.67619-0.00064-11.52503 DINC4 (75+) 0.000294.166970.000142.58220 DGEN2 (Female) 0.0019948.540940.0009629.24686 DPUR2 (Gifts) 0.0006817.049290.0011734.90372 DAGE2 (25/39) 0.000345.060800.000407.58485 DAGE3 (40/55) 0.0006710.075360.000428.01791 DAGE4 (55+) 0.0020130.448880.0010019.21276 DMTH2 (February) 0.000957.567040.0010711.20768 DMTH3 (March) 0.000352.765220.000000.00649 DMTH4 (April) 0.000826.470980.000121.17938 DMTH5 (May) 0.001239.808220.000889.07457 DMTH6 (June) -0.00054-4.05959-0.00018-1.71146 DMTH7 (July) -0.00085-6.37940-0.00030-2.86447 DMTH8 (August) -0.00072-5.41974-0.00023-2.17856 DMTH9 (September) -0.00061-4.61490-0.00023-2.17776 DMTH10(October) -0.00020-1.51145-0.00004-0.41734 DMTH11(November) -0.00032-2.46115-0.00022-2.15323 DMTH12(December) 0.000483.75451-0.00066-6.05477 SIGMA 0.0033297.135560.0023876.74114 MARKET PENETRATION WEST SOUTH CENTRAL INDOOR CUT-FLOWERS

PAGE 238

218 Table A.50 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in West South Central BETA T-VALUE BETA T-VALUE C -0.00295-33.77300-0.00132-29.50101 DINC2 (25/50) 0.000426.763320.0005110.21063 DINC3 (50/75) -0.00070-9.90934-0.00045-8.41078 DINC4 (75+) 0.000172.660940.000061.22673 DGEN2 (Female) 0.0006316.074180.0008325.88832 DPUR2 (Gifts) 0.0016629.680330.0007322.90012 DAGE2 (25/39) 0.000477.481330.000264.99058 DAGE3 (40/55) 0.000467.334260.000336.46593 DAGE4 (55+) 0.000497.874320.0009518.95089 DMTH2 (February) 0.000706.261370.0010411.46163 DMTH3 (March) 0.000090.72326-0.00006-0.60008 DMTH4 (April) -0.00015-1.168270.000181.85110 DMTH5 (May) 0.000655.820480.000788.36467 DMTH6 (June) -0.00021-1.68473-0.00013-1.32882 DMTH7 (July) -0.00028-2.21523-0.00027-2.63765 DMTH8 (August) -0.00018-1.44716-0.00019-1.84430 DMTH9 (September) -0.00020-1.59947-0.00017-1.70029 DMTH10(October) 0.000020.17291-0.00002-0.19480 DMTH11(November) -0.00012-0.98832-0.00022-2.19577 DMTH12(December) -0.00023-1.83292-0.00067-6.22689 SIGMA 0.0021647.415810.0022367.88253 MARKET PENETRATION WEST SOUTH CENTRAL ARRANGE NON-ARRAN

PAGE 239

219 Table A.51 Market Penetration Model I Results for Plants and Dry/Artificial in West South Central BETA T-VALUE BETA T-VALUE C -0.00143-28.49640-0.00220-35.36138 DINC2 (25/50) 0.0006811.582340.000407.54026 DINC3 (50/75) -0.00094-14.53618-0.00111-17.45052 DINC4 (75+) 0.000203.303120.000101.83017 DGEN2 (Female) 0.0015740.284390.0016037.26964 DPUR2 (Gifts) -0.00025-6.98939-0.00006-2.01740 DAGE2 (25/39) 0.000142.204660.000173.18325 DAGE3 (40/55) 0.0006611.004870.000468.51190 DAGE4 (55+) 0.0016027.071500.0011421.51855 DMTH2 (February) 0.000100.898950.000212.09235 DMTH3 (March) 0.000413.602190.000373.72446 DMTH4 (April) 0.001039.403280.000181.74009 DMTH5 (May) 0.000888.037300.000222.19863 DMTH6 (June) -0.00044-3.67925-0.00032-2.98409 DMTH7 (July) -0.00087-6.99010-0.00021-2.00151 DMTH8 (August) -0.00065-5.26693-0.00029-2.70805 DMTH9 (September) -0.00067-5.44506-0.00010-0.95924 DMTH10(October) -0.00020-1.67830-0.00007-0.69912 DMTH11(November) -0.00027-2.251890.000151.42997 DMTH12(December) 0.0013512.45138-0.00002-0.19577 SIGMA 0.0026575.320470.0020157.31765 MARKET PENETRATION WEST SOUTH CENTRAL PLANTS DRY

PAGE 240

220 Table A.52 Market Penetration Model I Resu lts for Outdoor in West South Central BETA T-VALUE C -0.00350-38.90502 DINC2 (25/50) 0.000818.50332 DINC3 (50/75) -0.00120-11.60093 DINC4 (75+) 0.000545.54936 DGEN2 (Female) 0.0021034.34111 DPUR2 (Gifts) -0.00210-34.35885 DAGE2 (25/39) 0.000232.28011 DAGE3 (40/55) 0.0013513.97118 DAGE4 (55+) 0.0025326.48546 DMTH2 (February) 0.000100.52051 DMTH3 (March) 0.0025115.00705 DMTH4 (April) 0.0041525.66005 DMTH5 (May) 0.0032619.91396 DMTH6 (June) 0.000734.05618 DMTH7 (July) -0.00095-4.78929 DMTH8 (August) -0.00192-8.88774 DMTH9 (September) -0.00084-4.22493 DMTH10(October) 0.000542.99519 DMTH11(November) -0.00172-8.07867 DMTH12(December) -0.00357-14.08631 SIGMA 0.0039273.22571 MARKET PENETRATION WEST SOUTH CENTRAL OUTDOOR

PAGE 241

221 Table A.53 Buyer Frequency Model I Results for Indoor and Cut-Flow ers in West South Central BETA T-VALUE BETA T-VALUE C -0.13405-0.849320.475550.92973 DINC2 (25/50) 0.017900.36733-0.26878-2.48240 DINC3 (50/75) -0.00200-0.043370.269913.43712 DINC4 (75+) -0.00748-0.152260.108521.84845 DGEN2 (Female) 0.340385.13723-0.07802-0.63390 DPUR2 (Gifts) -0.09766-2.63194-0.45622-3.10154 DAGE2 (25/39) 0.054811.36223-0.16572-2.58782 DAGE3 (40/55) 0.113111.93225-0.35426-3.64917 DAGE4 (55+) 0.315184.958100.357892.47190 DMTH2 (February) 0.072111.03915-0.13235-1.01528 DMTH3 (March) 0.197512.98471-0.07636-0.81678 DMTH4 (April) 0.005540.08145-0.03318-0.35812 DMTH5 (May) 0.177682.607340.242862.30935 DMTH6 (June) -0.08044-1.10364-0.15599-1.53681 DMTH7 (July) -0.10349-1.40794-0.02322-0.23591 DMTH8 (August) 0.097031.312910.228702.32807 DMTH9 (September) -0.13066-1.79250-0.01066-0.10759 DMTH10(October) -0.01733-0.25326-0.24128-2.47766 DMTH11(November) -0.15761-2.218500.031900.31584 DMTH12(December) 0.026860.394480.268682.10056 PRT -0.01345-7.91158-0.00689-4.02459 MILLS 0.061120.28353-0.87217-1.85599 SIGMA 1.3197266.629291.3803844.80834 CUT-FLOWERS INDOOR FREQUENCY WEST SOUTH CENTRAL

PAGE 242

222 Table A.54 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in West South Central BETA T-VALUE BETA T-VALUE C 0.049400.011021.337621.28049 DINC2 (25/50) -0.28235-0.64641-0.41813-2.21603 DINC3 (50/75) 0.211500.388620.348453.09677 DINC4 (75+) 0.118981.129050.123901.51098 DGEN2 (Female) 0.036230.06619-0.33506-1.49803 DPUR2 (Gifts) -0.36193-0.24507-0.65028-3.29184 DAGE2 (25/39) -0.01742-0.04982-0.31636-3.80146 DAGE3 (40/55) -0.29864-0.60594-0.49199-3.42363 DAGE4 (55+) -0.03283-0.066860.291411.03890 DMTH2 (February) 0.251980.46599-0.58811-2.22465 DMTH3 (March) -0.40670-1.893490.045840.39600 DMTH4 (April) 0.142620.61797-0.07664-0.63356 DMTH5 (May) 0.489050.98397-0.01163-0.06561 DMTH6 (June) 0.192550.67700-0.20872-1.66201 DMTH7 (July) -0.16565-0.543470.067560.53371 DMTH8 (August) -0.33348-1.208720.331702.61466 DMTH9 (September) 0.233051.09274-0.07346-0.57052 DMTH10(October) -0.38043-1.71941-0.25405-2.06362 DMTH11(November) -0.05210-0.238820.242931.71539 DMTH12(December) 0.478601.850400.343951.65676 PRT -0.00010-0.05628-0.01387-3.85543 MILLS -1.05529-0.43515-1.54862-1.81534 SIGMA 1.4994518.635661.5285339.73751 ARRANGE NON-ARRAN FREQUENCY WEST SOUTH CENTRAL

PAGE 243

223 Table A.55 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in West South Central BETA T-VALUE BETA T-VALUE C -2.00698-2.901800.676350.20149 DINC2 (25/50) 0.195251.39235-0.08981-0.21599 DINC3 (50/75) -0.24066-2.063130.170290.22737 DINC4 (75+) 0.132981.676270.207071.33077 DGEN2 (Female) 0.610422.613440.151860.12312 DPUR2 (Gifts) -0.37818-9.66964-0.07501-1.11269 DAGE2 (25/39) 0.219963.064610.088730.45428 DAGE3 (40/55) 0.264531.608250.133770.28099 DAGE4 (55+) 0.551812.50517-0.23318-0.29591 DMTH2 (February) -0.07797-0.655030.079610.36927 DMTH3 (March) 0.414363.664180.307981.13306 DMTH4 (April) 0.313832.09526-0.05431-0.24452 DMTH5 (May) 0.365752.533010.285871.33711 DMTH6 (June) -0.11742-0.89852-0.16856-0.67158 DMTH7 (July) -0.16434-1.07380-0.31659-1.44657 DMTH8 (August) -0.16006-1.130790.345811.19107 DMTH9 (September) -0.33143-2.31251-0.01986-0.09587 DMTH10(October) 0.065580.571500.255681.30543 DMTH11(November) -0.43974-3.47441-0.43927-2.13484 DMTH12(December) 0.562313.85552-0.35754-1.77700 PRT -0.01699-4.89675-0.05216-9.52484 MILLS 1.052211.79406-0.69351-0.33915 SIGMA 1.5884142.103482.3660139.96750 DRY PLANTS FREQUENCY WEST SOUTH CENTRAL

PAGE 244

224 Table A.56 Buyer Frequency Model I Results for Outdoor in West South Central BETA T-VALUE C -1.22418-2.33727 DINC2 (25/50) 0.111991.18657 DINC3 (50/75) -0.10872-1.21295 DINC4 (75+) 0.195302.66281 DGEN2 (Female) 0.498854.07949 DPUR2 (Gifts) -0.86715-6.83449 DAGE2 (25/39) 0.237172.96174 DAGE3 (40/55) 0.146011.09615 DAGE4 (55+) 0.327752.09857 DMTH2 (February) 0.247771.84731 DMTH3 (March) 1.098286.07207 DMTH4 (April) 1.400436.69537 DMTH5 (May) 1.134075.87331 DMTH6 (June) 0.499383.69424 DMTH7 (July) -0.36918-2.45947 DMTH8 (August) -0.47834-2.48029 DMTH9 (September) -0.25872-1.70828 DMTH10(October) 0.096440.71216 DMTH11(November) -0.68962-3.61283 DMTH12(December) -1.89514-5.47118 PRT -0.02640-6.68882 MILLS 0.707821.79850 SIGMA 1.9094352.86500 OUTDOOR FREQUENCY WEST SOUTH CENTRAL

PAGE 245

225 Table A.57 Market Penetration Model I Results for Indoor and Cut-Flowers in Mountain BETA T-VALUE BETA T-VALUE C -0.00098-13.59485-0.00344-33.24097 DINC2 (25/50) 0.0010110.202300.000605.59052 DINC3 (50/75) -0.00189-17.62195-0.00150-12.63724 DINC4 (75+) 0.000484.727590.000948.81255 DGEN2 (Female) 0.0020834.069680.0012418.63004 DPUR2 (Gifts) 0.0009415.923140.0022030.61421 DAGE2 (25/39) 0.000605.980140.000736.71483 DAGE3 (40/55) 0.0013213.357590.0013212.35873 DAGE4 (55+) 0.0016716.975150.000827.62296 DMTH2 (February) 0.001095.827970.001447.35276 DMTH3 (March) 0.000271.423710.000391.89723 DMTH4 (April) 0.000985.212100.000633.11865 DMTH5 (May) 0.001668.994610.001085.39284 DMTH6 (June) -0.00071-3.59566-0.00053-2.42503 DMTH7 (July) -0.00072-3.65464-0.00043-2.01111 DMTH8 (August) -0.00071-3.59019-0.00033-1.53239 DMTH9 (September) -0.00076-3.84115-0.00061-2.80702 DMTH10(October) -0.00076-3.86307-0.00059-2.71779 DMTH11(November) -0.00025-1.26171-0.00015-0.72852 DMTH12(December) 0.000603.16523-0.00044-2.02100 SIGMA 0.0047583.396010.0045162.85085 MARKET PENETRATION MOUNTAIN INDOOR CUT-FLOWERS

PAGE 246

226 Table A.58 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in Mountain BETA T-VALUE BETA T-VALUE C -0.00687-30.05508-0.00406-34.38989 DINC2 (25/50) 0.000846.260290.000464.17909 DINC3 (50/75) -0.00097-6.42540-0.00151-12.07513 DINC4 (75+) 0.000473.477390.000978.84199 DGEN2 (Female) 0.0008710.280990.0011116.08141 DPUR2 (Gifts) 0.0031623.285180.0015421.63384 DAGE2 (25/39) 0.000715.125210.000544.83443 DAGE3 (40/55) 0.001138.286890.0011210.18084 DAGE4 (55+) 0.000614.407170.000918.19364 DMTH2 (February) 0.000863.456560.001588.01297 DMTH3 (March) 0.000371.455020.000351.66863 DMTH4 (April) 0.000532.067000.000432.05498 DMTH5 (May) 0.000813.258440.001055.19544 DMTH6 (June) -0.00068-2.37962-0.00039-1.74262 DMTH7 (July) -0.00091-3.11638-0.00026-1.17797 DMTH8 (August) -0.00034-1.22131-0.00043-1.93986 DMTH9 (September) -0.00031-1.14053-0.00063-2.80264 DMTH10(October) -0.00062-2.18045-0.00037-1.69924 DMTH11(November) -0.00002-0.07507-0.00022-1.01078 DMTH12(December) 0.000281.06611-0.00062-2.73070 SIGMA 0.0042538.924530.0044254.75151 MARKET PENETRATION MOUNTAIN ARRANGE NON-ARRAN

PAGE 247

227 Table A.59 Market Penetration Model I Results for Plants and Dry/Artificial in Mountain BETA T-VALUE BETA T-VALUE C -0.00309-33.30344-0.00517-30.68334 DINC2 (25/50) 0.000879.321460.000736.85208 DINC3 (50/75) -0.00142-13.40209-0.00112-8.86976 DINC4 (75+) 0.000222.285340.000070.66649 DGEN2 (Female) 0.0018029.173480.0021323.16343 DPUR2 (Gifts) -0.00041-7.40692-0.00042-6.56975 DAGE2 (25/39) 0.000282.86529-0.00010-0.90921 DAGE3 (40/55) 0.000828.583150.000655.92153 DAGE4 (55+) 0.0018119.241420.0014813.70034 DMTH2 (February) 0.000422.35121-0.00022-1.00111 DMTH3 (March) 0.000251.405310.000211.03779 DMTH4 (April) 0.001206.969680.000110.54942 DMTH5 (May) 0.001559.108320.000532.65622 DMTH6 (June) -0.00062-3.23871-0.00018-0.85067 DMTH7 (July) -0.00073-3.75164-0.00021-0.98745 DMTH8 (August) -0.00093-4.71352-0.00004-0.21065 DMTH9 (September) -0.00071-3.66918-0.00010-0.45535 DMTH10(October) -0.00087-4.400250.000090.45236 DMTH11(November) -0.00023-1.21287-0.00003-0.13689 DMTH12(December) 0.001367.927350.000010.02696 SIGMA 0.0039461.086380.0035039.37398 MARKET PENETRATION MOUNTAIN PLANTS DRY

PAGE 248

228 Table A.60 Market Penetration Model I Results for Outdoor in Mountain BETA T-VALUE C -0.00639-39.22349 DINC2 (25/50) 0.001359.84678 DINC3 (50/75) -0.00112-7.50980 DINC4 (75+) 0.000251.73718 DGEN2 (Female) 0.0023526.28416 DPUR2 (Gifts) -0.00263-28.66022 DAGE2 (25/39) 0.000885.91916 DAGE3 (40/55) 0.0014810.13066 DAGE4 (55+) 0.0033723.42627 DMTH2 (February) -0.00084-2.90169 DMTH3 (March) 0.001285.00319 DMTH4 (April) 0.0032713.63359 DMTH5 (May) 0.0067129.67522 DMTH6 (June) 0.0033514.02519 DMTH7 (July) 0.000220.81968 DMTH8 (August) -0.00139-4.58706 DMTH9 (September) -0.00093-3.19161 DMTH10(October) -0.00077-2.67193 DMTH11(November) -0.00257-7.82317 DMTH12(December) -0.00526-12.27882 SIGMA 0.0051459.38247 MARKET PENETRATION MOUNTAIN OUTDOOR

PAGE 249

229 Table A.61 Buyer Frequency Model I Results for Indoor and CutFlowers in Mountain BETA T-VALUE BETA T-VALUE C -1.62322-2.79370-4.56136-2.71236 DINC2 (25/50) 0.283962.214300.282031.50217 DINC3 (50/75) -0.57315-4.28421-0.95341-3.35648 DINC4 (75+) 0.228432.907850.649844.56563 DGEN2 (Female) 0.483902.823010.642762.47850 DPUR2 (Gifts) 0.008290.098730.969462.23655 DAGE2 (25/39) 0.178652.164540.316011.89230 DAGE3 (40/55) 0.370942.505960.873632.90744 DAGE4 (55+) 0.326502.087500.524492.14250 DMTH2 (February) 0.173631.444080.490941.91913 DMTH3 (March) 0.210552.081060.280481.94915 DMTH4 (April) 0.343122.852860.392691.99192 DMTH5 (May) 0.624344.881890.682223.62191 DMTH6 (June) -0.13470-1.09006-0.20378-1.12917 DMTH7 (July) -0.17297-1.43579-0.04282-0.28442 DMTH8 (August) -0.06849-0.59935-0.29189-1.91103 DMTH9 (September) -0.17144-1.48643-0.31845-1.93955 DMTH10(October) -0.28583-2.38975-0.12421-0.71613 DMTH11(November) -0.39614-3.54126-0.39751-2.61494 DMTH12(December) 0.084130.80143-0.31381-1.79889 PRT -0.01225-5.24457-0.00834-3.42121 MILLS 1.175461.923752.580302.08400 SIGMA 1.7074551.542271.6245632.25526 INDOOR CUT-FLOWERS FREQUENCY MOUNTAIN

PAGE 250

230 Table A.62 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in Mountain BETA T-VALUE BETA T-VALUE C -11.38683-0.90144-7.88490-2.22851 DINC2 (25/50) 0.938340.819770.380651.27557 DINC3 (50/75) -1.07936-0.99261-1.61750-2.62187 DINC4 (75+) 0.523541.189021.039283.15526 DGEN2 (Female) 0.871630.867250.971382.00794 DPUR2 (Gifts) 3.298630.864781.162421.84860 DAGE2 (25/39) 0.612760.786010.578512.26958 DAGE3 (40/55) 1.129990.750371.242612.55255 DAGE4 (55+) 0.509180.484961.070752.09443 DMTH2 (February) 0.519500.487500.989511.72521 DMTH3 (March) 0.595000.998810.438402.17736 DMTH4 (April) 0.573780.820290.538802.08312 DMTH5 (May) 1.277121.562301.037162.70770 DMTH6 (June) -0.85055-0.87712-0.20832-0.89457 DMTH7 (July) -0.52836-0.463840.015120.08186 DMTH8 (August) -0.12722-0.25222-0.49430-1.94615 DMTH9 (September) -0.61002-1.40429-0.51051-1.90266 DMTH10(October) -0.47989-0.66917-0.07357-0.35856 DMTH11(November) -0.27173-0.82390-0.49353-2.36815 DMTH12(December) 0.252980.75087-0.98773-2.90276 PRT 0.000640.21913-0.00828-1.74211 MILLS 4.012210.657454.410441.84589 SIGMA 1.8642213.410451.8239928.40301 ARRANGE NON-ARRAN FREQUENCY MOUNTAIN

PAGE 251

231 Table A.63 Buyer Frequency Model I Results fo r Plants and Dry/Artificial in Mountain BETA T-VALUE BETA T-VALUE C -0.28271-0.1402221.379201.44994 DINC2 (25/50) -0.08976-0.30215-2.00486-1.43871 DINC3 (50/75) -0.09169-0.264132.594811.35614 DINC4 (75+) 0.297912.449100.433581.86759 DGEN2 (Female) -0.31291-0.64540-5.30307-1.34661 DPUR2 (Gifts) -0.40755-3.771071.328731.82289 DAGE2 (25/39) 0.233581.84698-0.00876-0.03358 DAGE3 (40/55) -0.18181-0.61980-2.05795-1.62828 DAGE4 (55+) -0.30178-0.56226-4.42004-1.59223 DMTH2 (February) -0.06013-0.281530.833521.28403 DMTH3 (March) 0.116960.65468-0.53613-0.90335 DMTH4 (April) 0.198160.614240.136910.31887 DMTH5 (May) 0.496811.36396-0.31558-0.33702 DMTH6 (June) 0.289101.25454-0.04325-0.07994 DMTH7 (July) -0.40213-1.450910.549481.21748 DMTH8 (August) 0.156850.561230.808021.80351 DMTH9 (September) -0.10461-0.427860.030580.07832 DMTH10(October) -0.39575-1.31224-0.45247-1.09043 DMTH11(November) -0.33863-1.73722-0.78224-1.89701 DMTH12(December) 0.305140.95686-0.88000-2.11885 PRT -0.01964-3.69589-0.03992-4.77291 MILLS -0.62284-0.42878-11.61578-1.54847 SIGMA 1.9699932.464513.1883024.69606 PLANTS DRY FREQUENCY MOUNTAIN

PAGE 252

232 Table A.64 Buyer Frequency Model I Results for Outdoor in Mountain BETA T-VALUE C -2.29309-2.20095 DINC2 (25/50) 0.301481.95293 DINC3 (50/75) -0.12078-1.00176 DINC4 (75+) 0.159071.37588 DGEN2 (Female) 0.438942.45578 DPUR2 (Gifts) -1.28792-5.89847 DAGE2 (25/39) 0.439353.22332 DAGE3 (40/55) 0.481862.74569 DAGE4 (55+) 0.454961.50685 DMTH2 (February) -0.28388-1.26514 DMTH3 (March) 0.403921.81791 DMTH4 (April) 1.131733.73026 DMTH5 (May) 2.193674.66047 DMTH6 (June) 1.266024.13496 DMTH7 (July) 0.148430.75786 DMTH8 (August) -0.76980-3.12028 DMTH9 (September) -0.12195-0.54911 DMTH10(October) 0.437582.07932 DMTH11(November) -0.57879-1.69062 DMTH12(December) -2.79782-3.98911 PRT -0.04152-7.24627 MILLS 1.007321.57061 SIGMA 2.2111443.35285 OUTDOOR FREQUENCY MOUNTAIN

PAGE 253

233 Table A.65 Market Penetration Model I Resu lts for Indoor and Cut-Flowers in Pacific BETA T-VALUE BETA T-VALUE C 0.0011631.409560.000010.53127 DINC2 (25/50) -0.00004-0.69149-0.00003-0.82460 DINC3 (50/75) -0.00057-9.32155-0.00019-4.52864 DINC4 (75+) 0.0010818.124320.0006516.18059 DGEN2 (Female) 0.0017950.570760.0008735.56775 DPUR2 (Gifts) 0.0004212.128150.0009237.16851 DAGE2 (25/39) 0.000457.595370.0004110.11088 DAGE3 (40/55) 0.0010217.287670.0008822.17270 DAGE4 (55+) 0.0016227.561640.0005714.22693 DMTH2 (February) 0.000645.620710.000689.00712 DMTH3 (March) 0.000363.136000.000283.60283 DMTH4 (April) 0.000635.521630.000344.45009 DMTH5 (May) 0.000998.790430.000668.81750 DMTH6 (June) -0.00033-2.84097-0.00012-1.52409 DMTH7 (July) -0.00052-4.44170-0.00031-3.95682 DMTH8 (August) -0.00045-3.88576-0.00021-2.66864 DMTH9 (September) -0.00060-5.06375-0.00025-3.13157 DMTH10(October) -0.00036-3.12150-0.00016-2.07732 DMTH11(November) -0.00029-2.51779-0.00025-3.17651 DMTH12(December) 0.000534.60328-0.00032-4.04135 SIGMA 0.00307108.017860.0019693.29890 MARKET PENETRATION PACIFIC INDOOR CUT-FLOWERS

PAGE 254

234 Table A.66 Market Penetration Model I Re sults for Flower Arrangements and NonArrangements in Pacific BETA T-VALUE BETA T-VALUE C -0.00180-35.16101-0.00020-7.47931 DINC2 (25/50) 0.000092.10489-0.00004-0.98156 DINC3 (50/75) -0.00005-1.12425-0.00015-3.87951 DINC4 (75+) 0.000358.495920.0005414.26334 DGEN2 (Female) 0.0003614.322510.0007934.13767 DPUR2 (Gifts) 0.0012034.186880.0006126.56827 DAGE2 (25/39) 0.000307.251970.000328.55802 DAGE3 (40/55) 0.0005513.282900.0007820.86511 DAGE4 (55+) 0.000296.864720.0005213.92499 DMTH2 (February) 0.000526.960800.000547.63052 DMTH3 (March) -0.00016-1.892300.000334.61239 DMTH4 (April) 0.000162.006190.000283.90809 DMTH5 (May) 0.000557.405130.000527.31731 DMTH6 (June) -0.00024-2.87229-0.00004-0.60919 DMTH7 (July) -0.00021-2.45359-0.00027-3.54218 DMTH8 (August) -0.00028-3.29909-0.00010-1.29952 DMTH9 (September) -0.00017-1.99264-0.00021-2.74216 DMTH10(October) -0.00008-1.03398-0.00012-1.59372 DMTH11(November) 0.000080.99805-0.00028-3.67217 DMTH12(December) 0.000182.25061-0.00042-5.50595 SIGMA 0.0015454.145110.0018387.98503 MARKET PENETRATION PACIFIC ARRANGE NON-ARRAN

PAGE 255

235 Table A.67 Market Penetration Model I Results for Plants and Dry/Artificial in Pacific BETA T-VALUE BETA T-VALUE C -0.00049-13.62963-0.00192-35.10733 DINC2 (25/50) 0.000050.981880.000143.30028 DINC3 (50/75) -0.00039-7.59900-0.00017-3.67398 DINC4 (75+) 0.0006913.864090.000102.16152 DGEN2 (Female) 0.0014446.399540.0011932.97302 DPUR2 (Gifts) -0.00032-10.94863-0.00039-14.74162 DAGE2 (25/39) 0.000275.295350.000102.12594 DAGE3 (40/55) 0.0006713.395640.000419.26499 DAGE4 (55+) 0.0016333.029020.0007617.37604 DMTH2 (February) 0.000131.378810.000212.56587 DMTH3 (March) 0.000161.625130.000101.19025 DMTH4 (April) 0.000626.626040.000010.12852 DMTH5 (May) 0.000636.776120.000040.43932 DMTH6 (June) -0.00018-1.84072-0.00043-4.57095 DMTH7 (July) -0.00041-4.11384-0.00025-2.82274 DMTH8 (August) -0.00039-3.94637-0.00013-1.44070 DMTH9 (September) -0.00050-5.02392-0.00011-1.25473 DMTH10(October) -0.00042-4.177710.000151.81242 DMTH11(November) -0.00021-2.113390.000161.93988 DMTH12(December) 0.0011011.926010.000101.22290 SIGMA 0.0023789.213580.0016152.84382 MARKET PENETRATION PACIFIC PLANTS DRY

PAGE 256

236 Table A.68 Market Penetration Model I Results for Outdoor in Pacific BETA T-VALUE C -0.00183-29.20525 DINC2 (25/50) 0.000303.71450 DINC3 (50/75) -0.00040-4.91109 DINC4 (75+) 0.0009912.42234 DGEN2 (Female) 0.0019639.82314 DPUR2 (Gifts) -0.00166-34.38599 DAGE2 (25/39) 0.000172.02416 DAGE3 (40/55) 0.0011414.15428 DAGE4 (55+) 0.0029737.38773 DMTH2 (February) 0.000060.42091 DMTH3 (March) 0.0015910.97449 DMTH4 (April) 0.0023316.31609 DMTH5 (May) 0.0030421.83322 DMTH6 (June) 0.001288.68018 DMTH7 (July) 0.000211.35841 DMTH8 (August) -0.00057-3.54036 DMTH9 (September) -0.00044-2.79914 DMTH10(October) -0.00122-7.33622 DMTH11(November) -0.00215-12.10070 DMTH12(December) -0.00247-13.61477 SIGMA 0.0035988.54833 MARKET PENETRATION PACIFIC OUTDOOR

PAGE 257

237 Table A.69 Buyer Frequency Model I Results for Indoor and CutFlowers in Pacific BETA T-VALUE BETA T-VALUE C 0.139251.69930-0.86989-4.63532 DINC2 (25/50) -0.06117-2.29008-0.15902-3.97655 DINC3 (50/75) -0.02986-1.144410.108802.93351 DINC4 (75+) 0.144155.509210.287167.61213 DGEN2 (Female) 0.185224.909070.290444.64995 DPUR2 (Gifts) -0.18509-7.30053-0.00508-0.07778 DAGE2 (25/39) -0.02563-0.747910.012140.24821 DAGE3 (40/55) 0.139993.552530.435705.87156 DAGE4 (55+) 0.158953.866570.389676.00165 DMTH2 (February) -0.00277-0.056380.119431.60342 DMTH3 (March) 0.070491.463270.088361.26831 DMTH4 (April) 0.108752.267380.174782.53113 DMTH5 (May) 0.182573.776740.368865.29502 DMTH6 (June) -0.01288-0.257490.119531.67508 DMTH7 (July) -0.10509-2.01921-0.19875-2.60781 DMTH8 (August) 0.046650.93615-0.02400-0.33007 DMTH9 (September) -0.04624-0.87621-0.05123-0.68379 DMTH10(October) -0.09911-1.95317-0.16394-2.21077 DMTH11(November) -0.01816-0.36200-0.22187-2.94602 DMTH12(December) -0.03658-0.746270.025830.33775 PRT -0.00988-7.68254-0.00903-5.65920 MILLS -0.04527-0.302700.720553.06913 SIGMA 1.0643177.586501.2785856.71970 CUT-FLOWERS INDOOR FREQUENCY PACIFIC

PAGE 258

238 Table A.70 Buyer Frequency Model I Resu lts for Flower Arrangements and NonArrangements in Pacific BETA T-VALUE BETA T-VALUE C -6.86863-2.25623-1.06816-3.31485 DINC2 (25/50) 0.113800.63344-0.16889-3.49359 DINC3 (50/75) 0.123281.171370.134933.22045 DINC4 (75+) 0.530652.364580.302436.81476 DGEN2 (Female) 0.732752.429140.311323.13479 DPUR2 (Gifts) 1.806101.67523-0.06623-0.80002 DAGE2 (25/39) 0.304381.124040.014250.24095 DAGE3 (40/55) 0.984971.830950.503614.50467 DAGE4 (55+) 0.428431.168850.470354.94144 DMTH2 (February) 0.844851.940140.081640.87651 DMTH3 (March) -0.37299-1.303870.134401.67261 DMTH4 (April) 0.197190.912050.196382.48734 DMTH5 (May) 0.955582.310410.368714.32755 DMTH6 (June) -0.12864-0.493060.144871.81560 DMTH7 (July) 0.163780.71398-0.22628-2.52018 DMTH8 (August) -0.88918-2.922280.018270.22716 DMTH9 (September) -0.02928-0.13604-0.03557-0.41432 DMTH10(October) -0.59347-2.47442-0.13948-1.67847 DMTH11(November) -0.32449-1.36064-0.20321-2.30076 DMTH12(December) 0.575792.98015-0.08048-0.85214 PRT 0.000180.09888-0.01949-6.26667 MILLS 2.913221.632030.827982.27057 SIGMA 1.6490019.571531.3620252.61613 ARRANGE NON-ARRAN FREQUENCY PACIFIC

PAGE 259

239 Table A.71 Buyer Frequency Model I Results for Plants and Dry/Artificial in Pacific BETA T-VALUE BETA T-VALUE C -0.67918-1.91647-1.56882-0.38838 DINC2 (25/50) 0.008990.182690.080840.30264 DINC3 (50/75) -0.06550-1.64720-0.29035-1.60073 DINC4 (75+) 0.239086.096850.091640.78715 DGEN2 (Female) 0.277082.067910.834660.59520 DPUR2 (Gifts) -0.34871-14.34774-0.21039-0.53726 DAGE2 (25/39) 0.011480.163270.241851.23875 DAGE3 (40/55) 0.171741.55809-0.01620-0.02925 DAGE4 (55+) 0.237951.664910.373090.40750 DMTH2 (February) -0.08770-1.153120.375851.33448 DMTH3 (March) 0.131511.798010.185820.83088 DMTH4 (April) 0.299153.538180.027380.12524 DMTH5 (May) 0.178372.168710.179000.82007 DMTH6 (June) -0.19060-2.43226-0.42499-0.91913 DMTH7 (July) -0.07180-0.837340.394671.19374 DMTH8 (August) -0.01283-0.151320.280551.23057 DMTH9 (September) -0.12415-1.46358-0.28735-1.14855 DMTH10(October) -0.06578-0.81051-0.10628-0.44143 DMTH11(November) -0.03251-0.41889-0.11868-0.54380 DMTH12(December) 0.146881.62979-0.19659-0.74996 PRT -0.01458-6.30731-0.03505-7.68573 MILLS 0.330800.873840.642850.27708 SIGMA 1.2669853.858762.4495836.55230 DRY PLANTS FREQUENCY PACIFIC

PAGE 260

240 Table A.72 Buyer Frequency Model I Results for Outdoor in Pacific BETA T-VALUE C -0.57858-2.22033 DINC2 (25/50) 0.035680.62181 DINC3 (50/75) 0.139302.80605 DINC4 (75+) 0.127162.51975 DGEN2 (Female) 0.400195.45244 DPUR2 (Gifts) -0.76901-12.96237 DAGE2 (25/39) 0.078911.31354 DAGE3 (40/55) 0.282063.28680 DAGE4 (55+) 0.429233.46723 DMTH2 (February) -0.12719-1.30636 DMTH3 (March) 0.561405.48621 DMTH4 (April) 0.607295.53388 DMTH5 (May) 0.747215.94384 DMTH6 (June) 0.390064.03108 DMTH7 (July) 0.078610.83038 DMTH8 (August) -0.03379-0.32842 DMTH9 (September) -0.02617-0.26311 DMTH10(October) -0.14025-1.24582 DMTH11(November) -0.78904-5.41318 DMTH12(December) -0.78610-4.94019 PRT -0.02258-7.58254 MILLS 0.709792.79013 SIGMA 1.6791664.55192 OUTDOOR FREQUENCY PACIFIC

PAGE 261

241 APPENDIX B TSP PROGRAMS OPTIONS MEMORY=500; TITLE 'MARCO PALMA PH.D MARKET PEN ETRATION AND FREQ UENCY MODELS'; ? FLWFREQ#8.TSP MODEL I ; ?===============================================================; ? HERE WE ARE ADDING UP THE INVERSE MILLS RATIO TO THE FREQUENCY; ?===============================================================; ? OUT 'D:\ZSTUDENT\MPAL MA\TSPPRG\FLWFREQ'; ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\HWDV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\BUYV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\EXPV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\TRNV4.XLS'); ? OUT; ? IN 'D:\ZSTUDENT\MPALMA\TSPPRG\FLWFREQ'; IN 'c:\ZSTUDENT\MPALMA\TSPPRG\FLWFREQ'; HIST(DISCRETE) HWDPUR HWD AGE HWDGEN HWDINC; NYRSMTH=INT(HWDYRS*100 + HWDMTH); ? 200412 = YEAR 2004 AND DECEMBER FOR EXAMPLE; LIST ZHWDZ HWDIDD1 HWDIDD2 HWDIDD3 HWDYRS HWDMTH HWDPUR HWDAGE HWDGEN HWDINC HWDTOT_0 HWDTLT_0 HWDIND_0 HWDCUT_0 HWDCAG_0 HWDCNA_0 HWDPLT_0 HWDPFW_0 HWDPFO_0 HWDDRY_0 HWDOUT_0 HWDTTL_0 HWDTOT_1 HWDTLT_1 HWDIND_1 HWDCUT_1 HWDCAG_1 HWDCNA_1 HWDPLT_1 HWDPFW_1 HWDPFO_1 HWDDRY_1 HWDOUT_1 HWDTTL_1 HWDTOT_2 HWDTLT_2 HWDIND_2 HWDCUT_2 HWDCAG_2 HWDCNA_2 HWDPLT_2 HWDPFW_2 HWDPFO_2 HWDDRY_2 HWDOUT_2 HWDTTL_2 HWDTOT_3 HWDTLT_3 HWDIND_3 HWDCUT_3 HWDCAG_3 HWDCNA_3 HWDPLT_3 HWDPFW_3 HWDPFO_3 HWDDRY_3 HWDOUT_3 HWDTTL_3 HWDTOT_4 HWDTLT_4 HWDIND_4 HWDCUT_4 HWDCAG_4 HWDCNA_4 HWDPLT_4 HWDPFW_4 HWDPFO_4 HWDDRY_4 HWDOUT_4 HWDTTL_4 HWDTOT_5 HWDTLT_5 HWDIND_5 HWDCUT_5 HWDCAG_5 HWDCNA_5 HWDPLT_5 HWDPFW_5 HWDPFO_5 HWDDRY_5 HWDOUT_5 HWDTTL_5 HWDTOT_6 HWDTLT_6 HWDIND_6 HWDCUT_6 HWDCAG_6 HWDCNA_6 HWDPLT_6 HWDPFW_6 HWDPFO_6 HWDDRY_6 HWDOUT_6 HWDTTL_6 HWDTOT_7 HWDTLT_7 HWDIND_7 HWDCUT_7 HWDCAG_7 HWDCNA_7 HWDPLT_7 HWDPFW_7 HWDPFO_7 HWDDRY_7 HWDOUT_7 HWDTTL_7 HWDTOT_8 HWDTLT_8 HWDIND_8 HWDCUT_8 HWDCAG_8 HWDCNA_8 HWDPLT_8 HWDPFW_8 HWDPFO_8 HWDDRY_8 HWDOUT_8 HWDTTL_8 HWDTOT_9 HWDTLT_9 HWDIND_9 HWDCUT_9 HWDCAG_9 HWDCNA_9 HWDPLT_9 HWDPFW_9 HWDPFO_9 HWDDRY_9 HWDOUT_9 HWDTTL_9;

PAGE 262

242 LIST ZBUYZ BUYIDD1 BUYIDD2 BUYIDD3 BUYYRS BUYMTH BUYPUR BUYAGE BUYGEN BUYINC BUYTOT_0 BUYTLT_0 BUYIND_0 BUYCUT_0 BUYCAG_0 BUYCNA_0 BUYPLT_0 BUYPFW_0 BUYPFO_0 BUYDRY_0 BUYOUT_0 BUYTTL_0 BUYTOT_1 BUYTLT_1 BUYIND_1 BUYCUT_1 BUYCAG_1 BUYCNA_1 BUYPLT_1 BUYPFW_1 BUYPFO_1 BUYDRY_1 BUYOUT_1 BUYTTL_1 BUYTOT_2 BUYTLT_2 BUYIND_2 BUYCUT_2 BUYCAG_2 BUYCNA_2 BUYPLT_2 BUYPFW_2 BUYPFO_2 BUYDRY_2 BUYOUT_2 BUYTTL_2 BUYTOT_3 BUYTLT_3 BUYIND_3 BUYCUT_3 BUYCAG_3 BUYCNA_3 BUYPLT_3 BUYPFW_3 BUYPFO_3 BUYDRY_3 BUYOUT_3 BUYTTL_3 BUYTOT_4 BUYTLT_4 BUYIND_4 BUYCUT_4 BUYCAG_4 BUYCNA_4 BUYPLT_4 BUYPFW_4 BUYPFO_4 BUYDRY_4 BUYOUT_4 BUYTTL_4 BUYTOT_5 BUYTLT_5 BUYIND_5 BUYCUT_5 BUYCAG_5 BUYCNA_5 BUYPLT_5 BUYPFW_5 BUYPFO_5 BUYDRY_5 BUYOUT_5 BUYTTL_5 BUYTOT_6 BUYTLT_6 BUYIND_6 BUYCUT_6 BUYCAG_6 BUYCNA_6 BUYPLT_6 BUYPFW_6 BUYPFO_6 BUYDRY_6 BUYOUT_6 BUYTTL_6 BUYTOT_7 BUYTLT_7 BUYIND_7 BUYCUT_7 BUYCAG_7 BUYCNA_7 BUYPLT_7 BUYPFW_7 BUYPFO_7 BUYDRY_7 BUYOUT_7 BUYTTL_7 BUYTOT_8 BUYTLT_8 BUYIND_8 BUYCUT_8 BUYCAG_8 BUYCNA_8 BUYPLT_8 BUYPFW_8 BUYPFO_8 BUYDRY_8 BUYOUT_8 BUYTTL_8 BUYTOT_9 BUYTLT_9 BUYIND_9 BUYCUT_9 BUYCAG_9 BUYCNA_9 BUYPLT_9 BUYPFW_9 BUYPFO_9 BUYDRY_9 BUYOUT_9 BUYTTL_9; LIST ZEXPZ EXPIDD1 EXPIDD2 EXPIDD3 EXPYRS EXPMTH EXPPUR EXPAGE EXPGEN EXPINC EXPTOT_0 EXPTLT_0 EXPIND_0 EXPCUT_0 EXPCAG_0 EXPCNA_0 EXPPLT_0 EXPPFW_0 EXPPFO_0 EXPDRY_0 EXPOUT_0 EXPTTL_0 EXPTOT_1 EXPTLT_1 EXPIND_1 EXPCUT_1 EXPCAG_1 EXPCNA_1 EXPPLT_1 EXPPFW_1 EXPPFO_1 EXPDRY_1 EXPOUT_1 EXPTTL_1 EXPTOT_2 EXPTLT_2 EXPIND_2 EXPCUT_2 EXPCAG_2 EXPCNA_2 EXPPLT_2 EXPPFW_2 EXPPFO_2 EXPDRY_2 EXPOUT_2 EXPTTL_2 EXPTOT_3 EXPTLT_3 EXPIND_3 EXPCUT_3 EXPCAG_3 EXPCNA_3 EXPPLT_3 EXPPFW_3 EXPPFO_3 EXPDRY_3 EXPOUT_3 EXPTTL_3 EXPTOT_4 EXPTLT_4 EXPIND_4 EXPCUT_4 EXPCAG_4 EXPCNA_4 EXPPLT_4 EXPPFW_4 EXPPFO_4 EXPDRY_4 EXPOUT_4 EXPTTL_4 EXPTOT_5 EXPTLT_5 EXPIND_5 EXPCUT_5 EXPCAG_5 EXPCNA_5 EXPPLT_5 EXPPFW_5 EXPPFO_5 EXPDRY_5 EXPOUT_5 EXPTTL_5 EXPTOT_6 EXPTLT_6 EXPIND_6 EXPCUT_6 EXPCAG_6 EXPCNA_6 EXPPLT_6 EXPPFW_6 EXPPFO_6 EXPDRY_6 EXPOUT_6 EXPTTL_6 EXPTOT_7 EXPTLT_7 EXPIND_7 EXPCUT_7 EXPCAG_7 EXPCNA_7 EXPPLT_7 EXPPFW_7 EXPPFO_7 EXPDRY_7 EXPOUT_7 EXPTTL_7 EXPTOT_8 EXPTLT_8 EXPIND_8 EXPCUT_8 EXPCAG_8 EXPCNA_8 EXPPLT_8 EXPPFW_8 EXPPFO_8 EXPDRY_8 EXPOUT_8 EXPTTL_8 EXPTOT_9 EXPTLT_9 EXPIND_9 EXPCUT_9 EXPCAG_9 EXPCNA_9 EXPPLT_9 EXPPFW_9 EXPPFO_9 EXPDRY_9 EXPOUT_9 EXPTTL_9; LIST ZTRNZ TRNIDD1 TRNIDD2 TRNIDD3 TRNYRS TRNMTH TRNPUR TRNAGE TRNGEN TRNINC TRNTOT_0 TRNTLT_0 TRNIND_0 TRNCUT_0 TRNCAG_0 TRNCNA_0 TRNPLT_0 TRNPFW_0 TRNPFO_0 TRNDRY_0 TRNOUT_0 TRNTTL_0 TRNTOT_1 TRNTLT_1 TRNIND_1 TRNCUT_1 TRNCAG_1 TRNCNA_1 TRNPLT_1 TRNPFW_1 TRNPFO_1 TRNDRY_1 TRNOUT_1 TRNTTL_1 TRNTOT_2 TRNTLT_2 TRNIND_2 TRNCUT_2 TRNCAG_2 TRNCNA_2 TRNPLT_2 TRNPFW_2 TRNPFO_2 TRNDRY_2 TRNOUT_2 TRNTTL_2 TRNTOT_3 TRNTLT_3 TRNIND_3 TRNCUT_3 TRNCAG_3 TRNCNA_3 TRNPLT_3 TRNPFW_3 TRNPFO_3 TRNDRY_3 TRNOUT_3 TRNTTL_3

PAGE 263

243 TRNTOT_4 TRNTLT_4 TRNIND_4 TRNCUT_4 TRNCAG_4 TRNCNA_4 TRNPLT_4 TRNPFW_4 TRNPFO_4 TRNDRY_4 TRNOUT_4 TRNTTL_4 TRNTOT_5 TRNTLT_5 TRNIND_5 TRNCUT_5 TRNCAG_5 TRNCNA_5 TRNPLT_5 TRNPFW_5 TRNPFO_5 TRNDRY_5 TRNOUT_5 TRNTTL_5 TRNTOT_6 TRNTLT_6 TRNIND_6 TRNCUT_6 TRNCAG_6 TRNCNA_6 TRNPLT_6 TRNPFW_6 TRNPFO_6 TRNDRY_6 TRNOUT_6 TRNTTL_6 TRNTOT_7 TRNTLT_7 TRNIND_7 TRNCUT_7 TRNCAG_7 TRNCNA_7 TRNPLT_7 TRNPFW_7 TRNPFO_7 TRNDRY_7 TRNOUT_7 TRNTTL_7 TRNTOT_8 TRNTLT_8 TRNIND_8 TRNCUT_8 TRNCAG_8 TRNCNA_8 TRNPLT_8 TRNPFW_8 TRNPFO_8 TRNDRY_8 TRNOUT _8 TRNTTL_8 TRNTOT_9 TRNTLT_9 TRNIND_9 TRNCUT_9 TRNCAG_9 TRNCNA_9 TRNPLT_9 TRNPFW_9 TRNPFO_9 TRNDRY_9 TR NOUT_9 TRNTTL_9; ? STATEMENT TO REMOVE THE TOTAL FOR THE DEMOGRAPHICS AND THE OBSERVATIONS END WITH JUNE OF 2004; DD=(HWDPUR>0)& (HWDAGE>0)& (HWDGEN>0)& (HWDINC>0) & (NYRSMTH<200407); PRINT @NOB; HIST(DISCRETE) DD; SELECT DD=1; DOT(CHAR=%) tlt_ ind_ cut_ cag_ cna_ plt_ pfw_ pfo_ dry_ out_ ttl_; DOT(CHAR=#) 0-9; SELECT DD=1 & BUY.%.#>0; WW= BUY.%.#; FRQ.%.# = TRN.%.# / [BUY.%.#]; ? FREQUENCY PER BUYER WW= TRN.%.#; PRT.%.# = EX P.%.# / [TRN.%.#]; ? EXPENDITURES PER TRANSACTIONS WW= BUY.%.#; PRB.%.# = EXP.%.# / [BUY.% .#]; ? AVERAGE PRICE PAID BY BUYERS SELECT DD=1; PEN.%.# = BUY.%.# / HWD.%.#; ? MARKET PENETRATION MSD(BYVAR) BUY.%.# FRQ.%.# PEN.%.#; HIST(NBINS=20) BUY.%.#; DBUY.%.#= BUY.%.#>0; HIST(DISCRETE,NBINS=20) DBUY.%.#; HIST(NBINS=20) FRQ.%.#; DFRQ.%.#= (FRQ.%.#=1) + (FRQ.%.#>1)*2 ; HIST(DISCRETE,NBINS=20) DFRQ.%.#; HIST(NBINS=20) PEN.%.#; DPEN.%.#= PEN.%.#>0; HIST(DISCRETE,NBINS=20) DPEN.%.#; ENDDOT; ENDDOT; SELECT DD=1; ?========================================================; ? DUMMY FOR GENDER 1=MALE AND 2=FEMALE; ?========================================================; HIST(DISCRETE) HWDGEN ; DUMMY HWDGEN; DGEN2 = HWDGEN2-HWDGEN1; ?========================================================; ? DUMMY FOR PURPOSE 1=SELF AND 2=GIFT; ?========================================================; HIST(DISCRETE) HWDPUR ; DUMMY HWDPUR; DPUR2 = HWDPUR2-HWDPUR1;

PAGE 264

244 ?========================================================; ? DUMMY FOR AGE 1=UNDER 25, 2=25/39, 3=40/54, 4=55+; ?========================================================; HIST(DISCRETE) HWDAGE ; DUMMY HWDAGE; DOT 2-4; DAGE. = HWDAGE. HWDAGE1; ENDDOT; ?========================================================; ? DUMMY FOR INC 1=UNDER 25, 2=25/39, 3=40/54, 4=55+; ?========================================================; HIST(DISCRETE) HWDINC ; DUMMY HWDINC; DOT 2-4; DINC. = HWDINC. HWDINC1; ENDDOT; ?========================================================; ? DUMMY FOR MONTHS 1=JAN ,2=FEB, .... ?========================================================; HIST(DISCRETE) HWDMTH; DUMMY HWDMTH; DOT 2-12; DMTH. = HWDMTH. HWDMTH1; ENDDOT; SELECT DD=1; MFORM(TYPE=GEN,NROW=50,NCOL=100) MCOEF=0; LIST XVARX DINC2-DINC4 DGEN2 DPUR2 DAGE2-DAGE4 DMTH2-DMTH12 ; ?========================================================; ? MARKET PENETRATION MODELS; ?========================================================; SELECT DD=1; SET I=-1; SET K= 0; SET G= 0; DOT(CHAR=%) tlt_ ind_ cut_ cag_ cna_ plt_ pfw_ pfo_ dry_ out_ ttl_; SET G=G+1; DOT(CHAR=#,VALUE=J) 0; ? RUN ONE TIME FOR EACH REGION FROM 0-9; SET I=I+2; SET K=K+2; PRT=PRT.%.#; OLSQ PEN.%.# C XVARX; PROBIT PEN.%.# C XVARX ; MILL.%.#=@MILLS; MAT NR=NROW(@COEF); DO L=1 TO NR; SET MCOEF(1,I)=G; SET MCOE F(2,I)=J; SET MCOEF(3,I)=1; SET LL=L+3; SET MCOEF(LL,I)=@COEF(L); SET MCOEF(1,K)=G; SET MCOE F(2,K)=J; SET MCOEF(3,K)=1; SET LL=L+3; SET MCOEF(LL,K)=@T(L); ENDDO;

PAGE 265

245 SET I=I+2; SET K=K+2; TOBIT PEN.%.# C XVARX; ? TRUNCATED AT ZERO; MAT NR=NROW(@COEF); DO L=1 TO NR; SET MCOEF(1,I)=G; SET MCOE F(2,I)=J; SET MCOEF(3,I)=2; SET LL=L+3; SET MCOEF(LL,I)=@COEF(L); SET MCOEF(1,K)=G; SET MCOE F(2,K)=J; SET MCOEF(3,K)=2; SET LL=L+3; SET MCOEF(LL,K)=@T(L); ENDDO; ENDDOT; ENDDOT; SET I=I+1; SET K=K+1; ?========================================================; ? BUYER FREQUENCY MODELS; ?========================================================; SET G=0; SELECT DD=1; DOT(CHAR=%) tlt_ ind_ cut_ cag_ cna_ plt_ pfw_ pfo_ dry_ out_ ttl_; SET G=G+1; DOT(CHAR=#,VALUE=J) 0; ? RUN ONE TIME FOR EACH REGION FROM 0-9; SET I=I+2; SET K=K+2; PRT=PRT.%.#; OLSQ FRQ.%.# C XVARX PRT; MAT NR=NROW(@COEF); DO L=1 TO NR; SET MCOEF(1,I)=G; SET MCOE F(2,I)=J; SET MCOEF(3,I)=1; SET LL=L+3; SET MCOEF(LL,I)=@COEF(L); SET MCOEF(1,K)=G; SET MCOE F(2,K)=J; SET MCOEF(3,K)=1; SET LL=L+3; SET MCOEF(LL,K)=@T(L); ENDDO; SET I=I+2; SET K=K+2; TFRQ.%.# = FRQ.%.# -1; TOBIT TFRQ.%.# C XVARX PRT MILL.%.# ; ? TRUNCATED AT ZERO; MAT NR=NROW(@COEF); DO L=1 TO NR; SET MCOEF(1,I)=G; SET MCOE F(2,I)=J; SET MCOEF(3,I)=2; SET LL=L+3; SET MCOEF(LL,I)=@COEF(L); SET MCOEF(1,K)=G; SET MCOE F(2,K)=J; SET MCOEF(3,K)=2; SET LL=L+3; SET MCOEF(LL,K)=@T(L); ENDDO; ENDDOT; ENDDOT; WRITE(FORMAT=EXCEL,FILE='c:\zstudent\MPalma\TSPPRG\OLS_TOBIT.xls') MCOEF ; END;

PAGE 266

246 OPTIONS MEMORY=1600; TITLE 'MARCO PALMA PH.D MARKET PEN ETRATION AND FREQ UENCY MODELS'; ? FLWFREQ#7.TSP M O D E L I I ; ? OUT 'D:\ZSTUDENT\MPAL MA\TSPPRG\FLWFREQ'; ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\HWDV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\BUYV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\EXPV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\TRNV4.XLS'); ? OUT; ? IN 'd:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; IN 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ?==================================================================; ? CREATING THE DATA ARRANGED WITH REGIONS AND FORMS IN THE VECTORS; ?==================================================================; ? OUT 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ? MAT NR=NROW(ALLFLW); ? FREQ NONE; ? SMPL 1,NR; ? UNMAKE ALLFLW ZNUM ZYRS ZMTH ZPUR ZINC ZAGE ZGEN ZFRM ZREG ZHWD ZBUY ZEXP ZTRN; ? OUT; ? DBLIST 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ? ? FORMS tlt=1 ind=2 cut=3 cag=4 cna=5 plt=6 pfw=7 pfo=8 dry=9 out=10 ttl=11; ? REGIONS 0 1 2 3 4 5 6 7 8 9; HIST(DISCRETE) ZFRM; DD= ZPUR>0 & ZGEN>0 & ZINC>0 & ZAGE>0 & ZREG>0 & ZNUM<200407; FF= (ZFRM=3 | ZFRM=6 |ZFRM=9 |ZFRM=10 ) & ZREG>0; SELECT DD=1; PEN=ZBUY/ZHWD; ? PENETRATION; DPEN=(PEN>0); SELECT DD=1 & ZBUY>0 & ZEXP>0 & ZTRN>0; FRQ=ZTRN/ZBUY; ? FREQUENCY OF BUYING; DFRQ=FRQ-1; PRT=ZEXP/ZTRN; ? PRICE PER TRANSACTION; SELECT DD=1; DOT ZMTH ZPUR ZINC ZAGE ZGEN ZFRM ZREG; DUMMY ; ENDDOT; DPUR2=ZPUR2-ZPUR1; DGEN2=ZGEN2-ZGEN1; DOT 2-4; DINC.= ZINC. ZINC1; DAGE. = ZAGE. ZAGE1; ENDDOT; DOT 2-12; DMTH. = ZMTH. ZMTH1; ENDDOT; DOT 2-9; DREG. = ZREG. ZREG1; ENDDOT; DOT 2-4; DFRM. = ZFRM. ZFRM1; ENDDOT; XGENPRT=DGEN2*PRT; XPURPRT=DPUR2*PRT; XGENINC=DGEN2*ZINC; MFORM(TYPE=GEN,NROW=50,NCOL=10) ZBETAZ=0;

PAGE 267

247 ?=========================================================================== ===; ? RUNNING THE TOBIT ACROSS CUT, FLOWERING, DRY AND OUTDOOR 3 6 9 & 10; ?=========================================================================== ===; DOT(VALUE=H) 3 6 9 10; SELECT DD=1 & ZFRM=H; ? ***select form =x to get the form we want***; PROBIT DPEN C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9; MILLSPROB=@MILLS; MAT BPENP=@COEF; MAT TPENP=@T; MAT NR=NROW(BPENP); DO I=1 TO NR; SET J=1; SET ZBETAZ(I,J)=BPENP(I); SET J=2; SET ZBETAZ(I,J)=TPENP(I); ENDDO; TOBIT PEN C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9; MAT BPENT=@COEF; MAT TPENT=@T; MAT NR=NROW(BPENT); DO I=1 TO NR; SET J=3; SET ZBETAZ(I,J)=BPENT(I); SET J=4; SET ZBETAZ(I,J)=TPENT(I); ENDDO; TOBIT DFRQ C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9 PRT MILLSPROB; MAT BFRQN=@COEF; MAT TFRQN=@T; MAT NR=NROW(BFRQN); DO I=1 TO NR; SET J=5; SET ZBETAZ(I,J)=BFRQN(I); SET J=6; SET ZBETAZ(I,J)=TFRQN(I); ENDDO; TOBIT DFRQ C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9 PRT MILLSPROB XGENPRT XPURPRT ; MAT BFRQI=@COEF; MAT TFRQI=@T; MAT NR=NROW(BFRQI); DO I=1 TO NR; SET J=7; SET ZBETAZ(I,J)=BFRQI(I); SET J=8; SET ZBETAZ(I, J)=TFRQI(I); ENDDO; MAT ZBETAZ.=ZBETAZ; PRINT ZBETAZ.; ENDDOT; OUT 'C:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; DOT 3 6 9 10; KEEP ZBETAZ.; ENDDOT; OUT; DBLIST 'C:\ZSTUDENT\MP ALMA\TSPPRG\ALLFREQ'; MMAKE XXBETAXX ZBETAZ3 ZBETAZ6 ZBETAZ9 ZBETAZ10; ?WRITE(FORMAT=EXCEL,FILE='d:\ZSTUDENT \MPALMA\TSPPRG\BETA03.XLS') ZBETAZ; WRITE(FORMAT=EXCEL,FILE='c:\ZSTUDENT\M PALMA\TSPPRG\BETARE V.XLS') XXBETAXX; END;

PAGE 268

248 OPTIONS MEMORY=1600; TITLE 'MARCO PALMA PH.D MARKET PEN ETRATION AND FREQ UENCY MODELS'; ? FLWFREQ#7.TSP SIMULATIONS PROGRAM; ? OUT 'D:\ZSTUDENT\MPAL MA\TSPPRG\FLWFREQ'; ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\HWDV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\BUYV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\EXPV4.XLS'); ? READ(FORMAT=EXCEL,FILE='D:\ZSTUDENT\MPALMA\FLOWERDATA\TRNV4.XLS'); ? OUT; ? IN 'd:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; IN 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ?==================================================================; ? CREATING THE DATA ARRANGED WITH REGIONS AND FORMS IN THE VECTORS; ?==================================================================; ? OUT 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ? MAT NR=NROW(ALLFLW); ? FREQ NONE; ? SMPL 1,NR; ? UNMAKE ALLFLW ZNUM ZYRS ZMTH ZPUR ZINC ZAGE ZGEN ZFRM ZREG ZHWD ZBUY ZEXP ZTRN; ? OUT; ? DBLIST 'c:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; ? ? FORMS tlt=1 ind=2 cut=3 cag=4 cna=5 plt=6 pfw=7 pfo=8 dry=9 out=10 ttl=11; ? REGIONS 0 1 2 3 4 5 6 7 8 9; HIST(DISCRETE) ZFRM; DD= ZPUR>0 & ZGEN>0 & ZINC>0 & ZAGE>0 & ZREG>0 & ZNUM<200407; FF= (ZFRM=3 | ZFRM=6 |ZFRM=9 |ZFRM=10 ) & ZREG>0; SELECT DD=1; PEN=ZBUY/ZHWD; ? PENETRATION; DPEN=(PEN>0); SELECT DD=1 & ZBUY>0 & ZEXP>0 & ZTRN>0; FRQ=ZTRN/ZBUY; ? FREQUENCY OF BUYING; DFRQ=FRQ-1; ? TRUNCATED SINCE THE MINIMUM FREQUENCY IS 1; PRT=ZEXP/ZTRN; ? PRICE PER TRANSACTION; SELECT DD=1 & FF=1 & DPEN>0; HIST PEN; HIST(DISCRETE) DPEN; SELECT DD=1; ?====================================; ? MEAN PRICE FOR EACH TYPE OF FLOWERS; ?====================================; DOT(VALUE=H) 3 6 9 10; SELECT DD=1 & ZFRM=H; MSD(NOPRINT) PRT; SET MPRT.=@MEAN; SELECT DD=1; ENDDOT;

PAGE 269

249 SELECT DD=1; DOT ZMTH ZPUR ZINC ZAGE ZGEN ZFRM ZREG; DUMMY ; HIST(DISCRETE) .; ENDDOT; DPUR2=ZPUR2-ZPUR1; DGEN2=ZGEN2-ZGEN1; DOT 2-4; DINC.= ZINC. ZINC1; DAGE. = ZAGE. ZAGE1; ENDDOT; DOT 2-12; DMTH. = ZMTH. ZMTH1; ENDDOT; DOT 2-9; DREG. = ZREG. ZREG1; ENDDOT; DOT 2-4; DFRM. = ZFRM. ZFRM1; ENDDOT; XGENPRT=DGEN2*PRT; XPURPRT=DPUR2*PRT; XGENINC=DGEN2*ZINC; MFORM(TYPE=GEN,NROW=50,NCOL=10) ZBETAZ=0; PROC XXXXX; ? HERE THE COEFFICIENTS ARE STORED AS MATRICES FOR EACH FORM; ?=========================================================================; ? RUNNING THE TOBIT ACROSS CUT, FLOWERING, DRY AND OUTDOOR 3 6 9 & 10; ?=========================================================================; DOT(VALUE=H) 3 6 9 10; SELECT DD=1 & ZFRM=H; ? ***select form =x to get the form we want***; PROBIT DPEN C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9; MILLSPROB=@MILLS; MAT BPENP=@COEF; MAT TPENP=@T; MAT NR=NROW(BPENP); DO I=1 TO NR; SET J=1; SET ZBETAZ(I,J)=BPENP(I); SET J=2; SET ZBETAZ(I,J)=TPENP(I); ENDDO; ? BASE TOBIT FOR PENETRATION; TOBIT PEN C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9; MAT BPENT=@COEF; MAT TPENT=@T; MAT NR=NROW(BPENT); DO I=1 TO NR; SET J=3; SET ZBETAZ(I,J)=BPENT(I); SET J=4; SET ZBETAZ(I,J)=TPENT(I); ENDDO; ? TOBIT FREQUENCY WITH PENETRATION MILLS; TOBIT DFRQ C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9 PRT MILLSPROB; MAT BFRQN=@COEF; MAT TFRQN=@T; MAT NR=NROW(BFRQN); DO I=1 TO NR; SET J=5; SET ZBETAZ(I,J)=BFRQN(I); SET J=6; SET ZBETAZ(I,J)=TFRQN(I); ENDDO; ? TOBIT FREQUENCY WITH MILLS AND INTERACTIONS;

PAGE 270

250 TOBIT DFRQ C DPUR2 DGEN2 DINC2-DINC4 DAGE2-DAGE4 DMTH2-DMTH12 DREG2-DREG9 PRT MILLSPROB XGENPRT XPURPRT ; MAT BFRQI=@COEF; MAT TFRQI=@T; MAT NR=NROW(BFRQI); DO I=1 TO NR; SET J=7; SET ZBETAZ(I,J)=BFRQI(I); SET J=8; SET ZBETAZ(I, J)=TFRQI(I); ENDDO; MAT ZBETAZ.=ZBETAZ; PRINT ZBETAZ.; ENDDOT; OUT 'd:\ZSTUDENT\MPALMA\TSPPRG\ALLFREQ'; DOT 3 6 9 10; KEEP ZBETAZ.; ENDDOT; OUT; DBLIST 'd:\ZSTUDENT\MP ALMA\TSPPRG\ALLFREQ'; ENDPROC XXXXX; ?========================================================================; ?========================================================================; ? CREATING THE SIMULATOR FOR THE TOBI T PENETRATION AND FREQUENCY MODELS; ?=========================================================================; SET INTCP = 1; LIST SIMVAR PUR GEN INC AGE MTH REG PRT MILLSPROB XGENPRT XPURPRT ; DOT SIMVAR; SET SIM_.=0; SET FLWTYPE=0; SET ADJ=1; ENDDOT; PROC INIT; ? INTIIALIZING THE SIMULATION VARIABLES TO ZERO; DOT SIMVAR; SET SIM_.=0; SET ADJ=1; ENDDOT; ENDPROC INIT; MFORM(TYPE=GEN,NROW=400,NCOL=10) MTOBITM=0; ?========================================================================; ? TOBIT SIMULATORS; ?========================================================================; PROC ZZSIMZZ; SET U=U+1; DOT(VALUE=J) 2 3 4; SET WAGE.=(SIM_AGE=J)-(SIM _AGE=1); ENDDOT; ? AGE LESS YOUNGEST; DOT(VALUE=J) 2; SET WGEN.=(SIM_GEN=J)-(SIM_GEN=1); ENDDOT; ? MALES MINUS FEMALES; DOT(VALUE=J) 2; SET WPUR.=(SIM_PUR=J)-(S IM_PUR=1); ENDDOT; ? GIFT MINUS SELF; DOT(VALUE=J) 2 3 4; SET WINC.=(SIM_INC=J)-(SIM _INC=1); ENDDOT; ? INC LESS YOUNGEST; DOT(VALUE=J) 2-12; SET WMTH.=(SIM_MTH=J)-(SIM_MTH=1); ENDDOT; ? MONTHS LESS JANUARY; DOT(VALUE=J) 2-9; SET WREG.=(SIM_REG=J)-(S IM_REG=1); ENDDOT; ? REGIONS LESS REG=1; SET WPRT = [(FLWTYPE=3)*MPRT3 + (FLWTYPE=6)*MPRT6 + (FLWTYPE=9)*MPRT9 + (FLWTYPE=10)*MPRT10]*ADJ; ? DENOTES THE FLWTYPE; DOT(VALUE=J) 2; SET WGENPRT.=[ (SIM_GEN=J)-(SIM_GEN=1) ]*WPRT; ENDDOT; ? MALES MINUS FEMALES X PRICE;

PAGE 271

251 DOT(VALUE=J) 2; SET WPURPRT.=[ (SIM_PUR=J)-(SIM_P UR=1) ]*WPRT; ENDDOT; ? GIFT MINUS SELF X PRICE; SET FLWHH=INT(FLWTYPE); DOT(VALUE=H) 3 6 9 10; IF FLWHH=H; THEN; DO; MMAKE BETA ZBETAZ.; ENDDO; ELSE; DO; SET FLWHH=FLWHH; ENDDO; ENDDOT; ? IN BETA COL1=PENETRATION PROBIT, CO L3=PENETRATION TOBIT, COL7=FREQUENCY TOBIT ESTIMATES; MAT NR=NROW(BETA); MFORM(TYPE=GEN,NROW=NR,NCOL=1) MM=0; DO I=1 TO NR; SET MM(I)=( BETA(I,1)^=0); ENDDO; MAT R=SUM(MM); MFORM(TYPE=GENERAL,NCOL=R,NCOL=1) B1=0; DO I=1 TO R; SET B1(I)=BETA(I,1); ENDDO; DO I=1 TO NR; SET MM(I)=( BETA(I,3)^=0); ENDDO; MAT R=SUM(MM); SET RR=R-1; MFORM(TYPE=GENERAL,N COL=RR,NCOL=1) B2=0; DO I=1 TO RR; SET B2(I)=BETA(I,3); ENDDO; SET SIGT2 = BETA(R,3); DO I=1 TO NR; SET MM(I)=( BETA(I,7)^=0); ENDDO; MAT R=SUM(MM); SET RR=R-1; MFORM(TYPE=GENERAL,N COL=RR,NCOL=1) B3=0; DO I=1 TO RR; SET B3(I)=BETA(I,7); ENDDO; SET SIGT3 = BETA(R,7); ? PRINT B1 B2 B3 SIGT2 SIGT3; SET L = 0; ? MUST CREATE THE MILLS VALUE FROM THE SIMULATED X VALUES; SET INTCP =1; MMAKE Z INTCP WPUR2 WGEN2 WINC2-WINC4 WAGE2-WAGE4 WMTH2-WMTH12 WREG2WREG9; MAT NNR=NROW(Z); MAT NNC=NCOL(Z); ? VARIABLES FOR THE PROBIT MODEL; MAT ZB1 = Z'B1; ? PROBIT VARIABLES AND COEFFICIENTS; SET WMILLS = NORM(ZB1) /CNORM(ZB1); SET PROB1=CNORM(ZB1); MAT ZB2 = Z'B2; ? PROBIT VARIABLES AND COEFFICIENTS; MMAKE X INTCP WPUR2 WGEN2 WINC2-WINC4 WAGE2-WAGE4 WMTH2-WMTH12 WREG2WREG9 WPRT WMILLS WGENPRT2 WPURPRT2; ? TOBIT FREQUENCY; MAT XB3 = X'B3; SET NORM_L2 = NORM[ ( ZB2 ) / ( SIGT2 ) ]; SET CNORM_L2 = CNORM[ ( ZB2 ) / ( SIGT2 ) ]; SET LAM2= NORM_L2/CNORM_L2;

PAGE 272

252 SET PENE1 = ( ZB2 + SIGT2*LAM2); SET PENE2 = CNORM_L2*PENE1 + (1-CNORM_L2)*L; ? NEED TO CHECK ON THE LOWER LIMIT IS ONE NOT ZERO; SET NORM_L3 = NORM[ ( XB3 ) / ( SIGT3 ) ]; SET CNORM_L3 = CNORM[ ( XB3 ) / ( SIGT3 ) ]; SET LAM3= NORM_L3/CNORM_L3; SET FREQ1 = ( XB3 + SIGT3*LAM3); SET FREQ2 = CNORM_L3*FREQ1 + (1-CNORM_L3)*L; SET MTOBITM(U,1)= SIMNUM; SET MTOBITM(U,2)= FLWTYPE; ? TYPE OF FLOWERS SET MTOBITM(U,3)= SIMTYPE; ? VARIABLE BEING SIMULATED; SET MTOBITM(U,4)= K; ? THE VARIABLE VALUES FOR EACH DUMMY; SET MTOBITM(U,5)= PROB1; ? PROBABILITY OF MARKET PENETRATION; SET MTOBITM(U,6)= PENE1; SET MTOBITM(U,7)= PENE2; SET MTOBITM(U,8)= FREQ1+1; SET MTOBITM(U,9)= FREQ2+1; SET MTOBITM(U,10)=WPRT; ENDPROC ZZSIMZZ; SET U=0; DOT(VALUE=G) 3 6 9 10; ? 3=CUT FLOWERS, 6=FLOWERING PLANTS, 9=DRY FLOWERS, 10=OUTDOOR FLOWERS; ?============================================================; ? SIMULATION =1 ; ? AVERAGE HOUSEHOLD ; ?============================================================; SET K =1; SET SIMNUM = 1; SET SIMTYPE = 0; INIT; SET FLWTYPE = G; ? 3 6 9 10; ZZSIMZZ; ?============================================================; ? SIMULATION =2 1=UNDER 25YRS 2=25/39YRS ; ? AVERAGE HOUSEHOLD 3=40/55YRS 4=55+ YRS ; ?============================================================; SET SIMNUM = 2; SET SIMTYPE = 1; INIT; SET FLWTYPE = G; DO F=1 TO 4; SET SIM_AGE=F; SET K =F; ZZSIMZZ; ENDDO;

PAGE 273

253 ?============================================================; ? SIMULATION =3 ; ? GENDER OF HOUSEHOLD HEAD (MALES=1 & FEMALES=2) ; ?============================================================; SET SIMNUM = 3; SET SIMTYPE = 2; INIT; SET FLWTYPE = G; ? 3 6 9 10; DO F=1 TO 2; SET SIM_GEN=F; SET K =F; ZZSIMZZ; ENDDO; ?============================================================; ? SIMULATION =4 ; ? PURPOSE OF BUYING (SELF=1 & GIFT=2) ; ?============================================================; SET SIMNUM = 4; SET SIMTYPE = 3; INIT; SET FLWTYPE = G; DO F=1 TO 2; SET SIM_PUR=F; SET K =F; ZZSIMZZ; ENDDO; ?============================================================; ? SIMULATION =5 1=UNDER $25,000 2=$25/$50,000 ; ? INCOME 3= $50/$75,000 4= $75,000+ ; ?============================================================; SET SIMNUM = 5; SET SIMTYPE = 4; INIT; SET FLWTYPE = G; DO F=1 TO 4; SET SIM_INC=F; SET K =F; ZZSIMZZ; ENDDO; ?============================================================; ? SIMULATION =6 ; ? MONTH 1=JAN, 2=FEB, 3=MAR, .... ; ?============================================================; SET SIMNUM = 6; SET SIMTYPE = 5; INIT; SET FLWTYPE = G; DO F=1 TO 12; SET SIM_MTH=F; SET K =F; ZZSIMZZ; ENDDO;

PAGE 274

254 ?=============================================================; ? SIMULATION =7 NINE REGIONS ?1=New England 2=Middle Atlantic 3=East North Central; ?4=West North Central 5=South Atlantic 6=East South Central; ?7=West South Central 8=Mountain 9=Pacific ; ?============================================================; SET SIMNUM = 7; SET SIMTYPE = 6; INIT; SET FLWTYPE = G; DO F=1 TO 9; SET SIM_REG=F; SET K =F; ZZSIMZZ; ENDDO; ?============================================================; ? SIMULATION =8 ; ? ADJUSTMENTS TO PRICE BASE ADJ=1 ; ?============================================================; SET SIMNUM = 8; SET SIMTYPE = 7; INIT; SET FLWTYPE = G; DO F=.20 TO 2.2 BY .20; SET ADJ=F; SET K =F; ZZSIMZZ; ENDDO; ?============================================================; ? SIMULATION =9 ; ? ADJUSTMENTS TO PRICE BASE ADJ=1 WITH GENDER ; ?============================================================; SET SIMNUM = 9; SET SIMTYPE = 8; INIT; SET FLWTYPE = G; DO FF=1 TO 2; SET SIM_GEN=FF; DO F=.20 TO 2.2 BY .20; SET ADJ=F; SET K =FF*10 + F; ZZSIMZZ; ENDDO; ENDDO; ?============================================================; ? SIMULATION =10 ; ? ADJUSTMENTS TO PRICE BASE ADJ=1 WITH PURPOSE ; ?============================================================;

PAGE 275

255 SET SIMNUM = 10; SET SIMTYPE = 9; INIT; SET FLWTYPE = G; DO FF=1 TO 2; SET SIM_PUR=FF; DO F=.20 TO 2.2 BY .20; SET ADJ=F; SET K =FF*10 + F; ZZSIMZZ; ENDDO; ENDDO; ?=================END SIMULATIONS============================; enddot; WRITE(FORMAT=EXCEL,FILE='c:\ZSTUDENT\MPA LMA\TSPPRG\TOBITSIM.XLS') MTOBITM; END;

PAGE 276

256 LIST OF REFERENCES Barten, A.P. 1977. The Systems of Consumer Demand Functions Approach: A Review. Econometrica. Vol. 45. Jan:23-51. Becker, W.A. 1993. Products, Services, and Consumer Perceptions of Service Quality in the Retail Floral Industry of Texas. PhD Dissertation. Texas A&M University. College Station, Texas. Behe, B.K. 1989. Floral Purchase Behavior of Pennsylvanians. PhD Dissertation. Pennsylvania State University. University Park, Pennsylvania. Behe, B.K., T.A. Prince, and H.K. Tayama. 1992a. Analysis of Consumer Purchases of Floral Products in Supermarkets. Hortscience. Vol. 27.:455-459. Behe, B.K., T.A. Prince, and H.K. Tayama. 1992b. Market Segmentation of Supermarket Floral Customers. Hortscience. Vol. 27.:455-459. Cornick, J., T.L. Cox, and B.W. Gould. Fluid Milk Purchases: A Multivariate Tobit Analysis. American Journal of Agricultural Economics. Vol. 76. No.1. Feb:74-82. Deaton, A., and Muellbauer, J. 1980. Economics and Consumer Behavior. Cambridge University Press. Cambridge, Mass. Economic Research Service. United States Department of Agriculture (USDA). 2003. Floriculture and Nursery Crops. Yearbook 2003. Washington D.C. Ehrenberg, A.S. 1988. Repeat-Buying, Facts, Theory and Applications. Oxford University Press. New York, New York. Fleck, F. 1981. Regularities of Market Penetrati on Processes Caused by Individual Consumer Behavior. Oelgeschlager, Gunn & Hain Publishers Inc. Cambridge, Mass. Franses, P.H. and R. Paap. 2001. Quantitative Models in Marketing Research. Cambridge University Press. Cambridge, Mass. Girapunthong, N. 2002. Demand Drivers for Fresh Cut Flowers and Their Substitutes: An Application of Household Expenditure Allocation Models. Ph.D. Dissertation. Food and Resource Economics Department. Un iversity of Florida. Gainesville, Florida.

PAGE 277

257 Goldberg, A.S. 1967. Functional Form and Utility: A Review of Consumer Demand. Theory. Social Systems Research Institu te, University of Wisconsin: Systems Formulation, Methodology and Policy Workshop Paper 6703. Green, W.H. 2000. Econometric Analysis. 4th Edition. Prentice Hall. Upper Saddle River, New Jersey. Gujarati, D.N. 1995. Basic Econometrics. 3rd Edition. McGraw-Hill, Inc. Hall, B.H. 1992. Time Series Processor Version 4.2. Reference Manual. Palo Alto, California. Hicks, J.R. 1946. Value and Capital. 2nd Edition. Oxford University Press. New York, New York. Hotteling, H. 1932. EdgeworthÂ’s Taxation Paradoxand the Nature of Demand and Supply Functions. Journal of Political Economy. Vol.40. Oct:577-616. Ipsos-NPD Group. 2004. National Panel Di ary Group, Inc. Chicago, Illinois. Johnson, S.R., Z.A. Hassan., and R.D. Green. 1984. Demand Systems Estimation. Iowa State University Press. Ames, Iowa. Miller, M. 1983. Commodity Sub-Sector Analysis of Cut Flower Industry. PhD. Dissertation. Food and Resource Economics Department. University of Florida. Gainesville, Florida. Nicholson, W. 1998. Microeconomic Theory: Basic Principles and Extensions. Dryden Press. Fort Worth, Texas. Parsons, L.J. and R.J. Schultz. 1976. Marketing Models and Econometric Research. North Holland Publishing Company. New York, New York. Phlips, L. 1974. Applied Consumption Analysis. Amsterdam: North Holland. Pindyck, R.S. and D.L. Rubinfeld. 2001. Microeconomics. 5th Edition. Prentice Hall. Upper Saddle River, New Jersey. Powell, A.A. 1974. Empirical Analysis of Demand Systems. Lexington Books. Lexington, Mass. Prais, S., and H. Houthakker. 1955. The Analysis of Family Budgets. Cambridge University Press. New York, New York. Rimal, A.P. 1998. Effect of Generic and Brand Promoti ons of Fresh Cut Flowers on the Use of Retail Flower Outlets. Ph.D. Dissertation. F ood And Resource Economics. University of Florida. Gainesville, Florida.

PAGE 278

258 Samuelson, P.A. 1947. Foundations of Economic Analysis. Harvard University Press. Cambridge, Mass. Schumacher, S.K., T.L. Marsh, and K.A. Williams. 2000. Economic Thresholds: An Application to Floriculture. Department of Agricultural Economics. Kansas State University. Manhattan, Kansas. Shephard, R.W. 1953. Cost and Production Functions. Princeton University Press. Princeton, New Jersey. Theil, H., and Gabrielson, A. 1975. Theory and Measurement of Consumer Demand. Vol. 1. Amsterdam: North Holland. Theil, H., and Gabrielson, A. 1976. Theory and Measurement of Consumer Demand. Vol. 2. Amsterdam: North Holland. Tilburg, A.V. 1984. Consumer Choice of Cut Flowers and Pot Plants. Agricultural University. Wageningen, The Netherlands. Tobin, J. 1958. Estimation of Relationships fo r Limited Dependent Variables. Econometrica. Vol.26.:24-36. United States Department of Agriculture (USDA). 2003. Floriculture and Nursery Crops. Yearbook 2003. Washington D.C. Ward, R.W. 1997. Evaluating Promoflor: Have the Promotions of Fresh Cut Flowers and Greens Had an Impact. National Promoflor Council. UF/PER97#2. University of Florida. Gainesville, Florida. Ward, R.W. 2004. Estimated the Impact of FPOÂ’s Generic Promotions of Fresh Cut Flowers: Analysis Through Phase VI. Research Paper. Food And Resource Economics. University of Flor ida. Gainesville, Florida.

PAGE 279

259 BIOGRAPHICAL SKETCH Marco Antonio Palma Garcia was born December 24, 1979, in Tegucigalpa, Honduras. He graduated from the Instituto Salesiano San Miguel in 1996. In 1999, he graduated from the Pan American School of Agriculture, Zamorano, where the philosophy of “learn by doing” is strongly emphasized. He tran sferred to University of Florida in January 2000, and completed his B achelor of Science degree in the Food and Resource Economics Department with a speci alization in agribusiness management. He also received a major in interdisciplinary st udies, with a specialization in agricultural science. Upon completion of his bachelor’s degree he continued his studies at the University of Florida as a graduate student to pursue his Master of Science and Doctor of Philosophy degree in ag ricultural economics.