DETERMINING THE EFFECTS OF OBSERVED AND UNOBSERVED HETEROGENEITY ON CONSUMER BRAND CHOICE

This paper investigates the effects of heterogeneity in consumer choice behaviour. Omitted consumer heterogeneity may lead to badly biased results, and wrong inferences concerning marketing strategies to follow. In this research we study the extent and the cause of this bias. We distinguish between observed and unobserved heterogeneity, by partialing out the effects of unmeasured heterogeneity and modelling it explicitly. The following questions will be addressed: What is unobserved heterogeneity and how much of it can be explained? How should heterogeneity be incorporated in consumer choice models? A hazard model is used for the analysis. The hazard model will yield patterns of switching among brands, as well as, the effect of marketing mix variables on brand choice and purchase timing. Differences between switchers and repeat purchasers are studied and the extent to which brand choice can be explained. Our model is estimated using scanner panel data. We find that it is important to include both observed and unobserved heterogeneity in order to obtain a better fit of the model. Our results show that it may be sufficient to only include observed heterogeneity to obtain unbiased parameter estimates. Including observed heterogeneity also reduces the aggregation or heterogeneity bias in the hazard function. © 1998 John Wiley & Sons, Ltd.

[1]  J. Heckman,et al.  Models for the Analysis of Labor Force Dynamics , 1982 .

[2]  Terry Elrod,et al.  Choice Map: Inferring a Product-Market Map from Panel Data , 1988 .

[3]  Füsun F. Gönül,et al.  Modeling Multiple Sources of Heterogeneity in Multinomial Logit Models: Methodological and Managerial Issues , 1993 .

[4]  Multiple ways of measuring brand loyalty , 1996 .

[5]  Albert C. Bemmaor,et al.  Models of purchase timing and models of brand choice: Outlook and issues , 1991 .

[6]  Naufel J. Vilcassim,et al.  Investigating Household Purchase Timing Decisions: A Conditional Hazard Function Approach , 1991 .

[7]  Pradeep K. Chintagunta,et al.  Estimating a Multinomial Probit Model of Brand Choice Using the Method of Simulated Moments , 1992 .

[8]  Terry Elrod,et al.  A Factor-Analytic Probit Model for Representing the Market Structure in Panel Data , 1995 .

[9]  Hjp Harry Timmermans,et al.  An Unconditional Competing Risk Hazard Model of Consumer Store-Choice Dynamics , 1996 .

[10]  Peter E. Rossi,et al.  A Bayesian Approach to Estimating Household Parameters , 1993 .

[11]  Joel L. Horowitz,et al.  Sampling, Specification, and Data Errors in Probabilistic Discrete-Choice Models , 1981 .

[12]  P. Bentler,et al.  Significance Tests and Goodness of Fit in the Analysis of Covariance Structures , 1980 .

[13]  Kenneth Hilton,et al.  Repeat-Buying: Theory and Applications. , 1973 .

[14]  Fred S. Zufryden,et al.  Adding Explanatory Variables to a Consumer Purchase Behavior Model: An Exploratory Study , 1980 .

[15]  Greg M. Allenby,et al.  Reassessing Brand Loyalty, Price Sensitivity, and Merchandising Effects on Consumer Brand Choice , 1995 .

[16]  M. Hannan,et al.  Social Dynamics: Models and Methods. , 1986 .

[17]  J. Kalbfleisch,et al.  The Statistical Analysis of Failure Time Data , 1980 .

[18]  J. Heckman,et al.  A Method for Minimizing the Impact of Distributional Assumptions in Econometric Models for Duration Data , 1984 .

[19]  James J. Heckman,et al.  Using Goodness of Fit and Other Criteria to Choose among Competing Duration Models: A Case Study of Hutterite Data , 1987 .

[20]  A. Taudes,et al.  Microdynamics of new product purchase: A model incorporating both marketing and consumer-specific variables , 1991 .

[21]  Kannan Srinivasan,et al.  Consumer Purchase Behavior in a frequently Bought Product Category: Estimation Issues and Managerial Insights from a Hazard Function Model with Heterogeneity , 1993 .

[22]  Frank M. Bass,et al.  Pooling Issues and Methods in Regression Analysis with Examples in Marketing Research , 1975 .

[23]  Pradeep K. Chintagunta.  Heterogeneity in nested logit models: An estimation approach and empirical results , 1992 .

[24]  D. B. Montgomery,et al.  Individual differences in response to consumer promotions , 1991 .

[25]  N. Kiefer.  Economic Duration Data and Hazard Functions , 1988 .

[26]  Frank M. Bass,et al.  Equilibrium Stochastic Choice and Market Penetration Theories: Derivations and Comparisons , 1976 .

[27]  Terry Elrod,et al.  Internal analysis of market structure: Recent developments and future prospects , 1991 .

[28]  Tony Lancaster,et al.  ECONOMETRIC METHODS FOR THE DURATION OF UNEMPLOYMENT , 1979 .

[29]  Joel H. Steckel,et al.  A Heterogeneous Conditional Logit Model of Choice , 1988 .

[30]  David C. Schmittlein,et al.  Analyzing Duration Times in Marketing: Evidence for the Effectiveness of Hazard Rate Models , 1993 .

[31]  J. Heckman.  Heterogeneity and State Dependence , 1981 .

[32]  J. M. Jones,et al.  Removing Heterogeneity Bias from Logit Model Estimation , 1988 .

[33]  Cheng Hsiao,et al.  Analysis of Panel Data , 1987 .

[34]  Dipak C. Jain,et al.  A Random-Coefficients Logit Brand-Choice Model Applied to Panel Data , 1994 .

[35]  Gary J. Russell,et al.  A Probabilistic Choice Model for Market Segmentation and Elasticity Structure , 1989 .

[36]  A. W. Kemp,et al.  The Dirichlet: A comprehensive model of buying behaviour , 1984 .

[37]  S. Reader,et al.  Unobserved Heterogeneity in Dynamic Discrete Choice Models , 1993 .

[38]  Richard B. Davies,et al.  Recent Developments in the Analysis of Movement and Recurrent Choice , 1984 .

[39]  Naufel J. Vilcassim,et al.  Modeling Purchase-Timing and Brand-Switching Behavior Incorporating Explanatory Variables and Unobserved Heterogeneity , 1991 .

[40]  Frank M. Bass,et al.  A Multibrand Stochastic Model Compounding Heterogeneous Erlang Timing and Multinomial Choice Processes , 1980, Oper. Res..

[41]  Peter S. Fader,et al.  Accounting for Heterogeneity and Nonstationarity in a Cross-Sectional Model of Consumer Purchase Behavior , 1993 .

[42]  Frank M. Bass,et al.  An Investigation into the Order of the Brand Choice Process , 1984 .

[43]  Pradeep K. Chintagunta,et al.  Investigating Heterogeneity in Brand Preferences in Logit Models for Panel Data , 1991 .

[44]  James J. Heckman,et al.  Longitudinal Analysis of Labor Market Data: Social science duration analysis , 1985 .

[45]  D. Cox.  Regression Models and Life-Tables , 1972 .