×

The benefits of incorporating utility dependencies in finite mixture probit models. (English) Zbl 1371.91061

Summary: We propose an application of a new finite mixture multinomial conditional probit (FM-MNCP) model that accommodates preference heterogeneity and explicitly accounts for utility dependencies between choice alternatives considering both local and background contrast effects. The latter is accomplished by using a one-factor structure for segment-specific covariance matrices allowing for nonzero off-diagonal covariance elements. We compare the model to a finite mixture multinomial independent probit (FM-MNIP) model that as well accommodates heterogeneity but assumes independence. That way, we address the potential benefits of a model that additionally accounts for dependencies over a model that accommodates heterogeneity only. Our model comparison is based on empirical data for smoothies and is assessed in terms of fit, holdout validation, and market share predictions. One of the main findings of our empirical study is that allowing for utility dependencies may counterbalance the effects of considering heterogeneity, and vice versa. Additional findings from a simulation study indicate that the FM-MNCP model outperforms the FM-MNIP model with respect to parameter recovery.

MSC:

91B16 Utility theory

Software:

bayesm
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Andrews RL, Ainslie A, Currim IS (2002a) An empirical comparison of logit choice models with discrete versus continuous representations of heterogeneity. J Mark Res 39:479-487 · doi:10.1509/jmkr.39.4.479.19124
[2] Andrews R, Ansari A, Currim I (2002b) Hierarchical Bayes versus finite mixture conjoint analysis models: a comparison of fit, prediction, and partworth recovery. J Mark Res 39:87-98 · doi:10.1509/jmkr.39.1.87.18936
[3] Andrews R, Currim I (2003) A comparison of segment retention criteria for finite mixture logit models. J Mark Res 40:235-243 · doi:10.1509/jmkr.40.2.235.19225
[4] Baur A, Klein R, Steinhardt C (2014) Model-based decision support for optimal brochure pricing: applying advanced analytics in the tour operating industry. OR Spectrum 36:557-584 · Zbl 1305.90241 · doi:10.1007/s00291-013-0338-3
[5] Bolduc D, Fortin B, Fournier M (1996) The impact of incentive policies on practice location of doctors: a multinomial probit analysis. J Labor Econ 14:703-732 · doi:10.1086/209828
[6] Boztuğ Y, Hildebrandt L, Raman K (2014) Detecting price thresholds in choice models using a semi-parametric approach. OR Spectrum 36:187-207 · Zbl 1290.91063
[7] Bunch D (1991) Estimability in the multinomial probit model. Transp Res B 25B:1-12 · doi:10.1016/0191-2615(91)90009-8
[8] Crabbe M, Jones B, Vandebroek M (2013) Comparing two-stage segmentation methods for choice data with a one-stage latent class choice analysis. Commun Stat Simul Comput 42:1188-1212 · Zbl 1347.62256 · doi:10.1080/03610918.2011.654035
[9] Daganzo C (1979) Multinomial probit: the theory and its applications to demand forecasting. Academic Press, New York · Zbl 0476.62090
[10] Dansie B (1985) Parameter estimability in the multinomial probit model. Transp Res B 19B:526-528 · doi:10.1016/0191-2615(85)90047-5
[11] DeSarbo W, Ramaswamy V, Cohen S (1995) Market segmentation with choice-based conjoint analysis. Mark Lett 6:137-147 · doi:10.1007/BF00994929
[12] Dhar R, Nowlis S, Sherman S (2000) Trying hard or hardly trying: an analysis of context effects in choice. J Consum Psychol 9:189-200 · doi:10.1207/S15327663JCP0904_1
[13] Elrod T, Keane M (1995) A factor analytic probit model for representing the market structure of panel data. J Mark Res 32:1-16 · doi:10.2307/3152106
[14] Fennell G, Allenby GM, Yang S, Edwards Y (2003) The effectiveness of demographic and psychographic variables for explaining brand and product category use. Quant Mark Econ 1:223-244 · doi:10.1023/A:1024686630821
[15] Greene W, Hensher D (2013) Revealing additional dimensions of preference heterogeneity in a latent class mixed multinomial logit model. Appl Econ 45:1897-1902 · doi:10.1080/00036846.2011.650325
[16] Grün B, Leisch F (2008) Identifiability of finite mixtures of multinomial logit models with varying and fixed effects. J Classif 25:225-247 · Zbl 1276.62021 · doi:10.1007/s00357-008-9022-8
[17] Gustafson, A.; Herrmann, A.; Huber, F.; Gustafson, A. (ed.); Herrmann, A. (ed.); Huber, F. (ed.), Conjoint analysis as an instrument of market research practice, 5-46 (2003), Berlin · doi:10.1007/978-3-540-24713-5_1
[18] Haaijer R, Wedel M, Vriens M, Wansbeek TJ (1998) Utility covariances and context effects in conjoint MNP models. Mark Sci 17:236-252 · doi:10.1287/mksc.17.3.236
[19] Haase K, Müller S (2015) Insights into clients’ choice in preventive health care facility location planning. OR Spectrum 37:273-291 · Zbl 1308.90017 · doi:10.1007/s00291-014-0367-6
[20] Huber J, Arora N, Johnson RM (1998) Capturing heterogeneity in consumer choices. ART Forum. American Marketing Association, Chicago
[21] Hausman J, Wise D (1978) A conditional probit model for qualitative choice: discrete decisions recognizing interdependence and heterogeneous preferences. Econometrica 46:403-429 · Zbl 0376.90005 · doi:10.2307/1913909
[22] Karniouchina EV, Moore WL, van der Rhee B, Verma R (2009) Issues in the use of ratings-based versus choice-based conjoint analysis in operations management research. Eur J Oper Res 197:340-348 · Zbl 1157.90450 · doi:10.1016/j.ejor.2008.05.029
[23] Keane M (1992) A note on identification in the multinomial probit model. J Bus Econ Stat 10:193-200
[24] Keane M, Wasi N (2013) Comparing alternative models of heterogeneity in consumer choice behavior. J Appl Econom 28:1018-1045
[25] Löffler S, Baier D (2015) Bayesian conjoint analysis in water park pricing: a new approach taking varying part worths for attribute levels into account. J Serv Sci Manag 8:46-56
[26] McCullough RP (2009) Comparing hierarchical Bayes and latent class choice: practical issues for sparse data sets. In: 2009 Sawtooth software conference proceedings, pp 273-284
[27] McLachlan G, Krishnan T (2007) The EM algorithm and extensions. Wiley, Hoboken · Zbl 0882.62012
[28] Moore WL (2004) A cross-validity comparison of rating-based and choice-based conjoint analysis models. Int J Res Mark 21:299-312 · doi:10.1016/j.ijresmar.2004.01.002
[29] Moore WL, Gray-Lee J, Louviere JJ (1998) A cross-validity comparison of conjoint analysis and choice models at different levels of aggregation. Mark Lett 9:195-208 · doi:10.1023/A:1007913100332
[30] Natter M, Feurstein M (2002) Real world performance of choice-based conjoint models. Eur J Oper Res 137:448-458 · Zbl 1012.90519 · doi:10.1016/S0377-2217(01)00147-3
[31] Otter T, Tüchler R, Frühwirth-Schnatter S (2004) Capturing consumer heterogeneity in metric conjoint analysis using Bayesian mixture models. Int J Res Mark 21:285-297 · doi:10.1016/j.ijresmar.2003.11.002
[32] Özkan C, Karaesmen F, Özekici S (2015) A revenue management problem with a choice model of consumer behavior. OR Spectrum 37:457-473 · Zbl 1312.91065 · doi:10.1007/s00291-014-0380-9
[33] Paetz F (2013) Finite Mixture Multinomiales Probitmodell: Konzeption und Umsetzung. Springer, Wiesbaden · doi:10.1007/978-3-658-02662-2
[34] Paetz F, Steiner W (2014) A finite mixture multinomial probit model for choice based conjoint analysis: a simulation study. In: Proceedings of the 43th EMAC conference, University of Valencia, Spain
[35] Paetz F, Steiner W (2015) Die Berücksichtigung von Abhängigkeiten zwischen Alternativen in Finite Mixture Conjoint Choice Modellen: Eine Simulationsstudie. Mark ZFP 37:90-100 · doi:10.15358/0344-1369-2015-2-90
[36] Raftery A (1995) Bayesian model selection in social research. Sociol Methodol 25:111-163 · doi:10.2307/271063
[37] Rooderkerk R, Van Heerde H, Bijmolt T (2011) Incorporating context effects into a choice model. J Mark Res 48:767-780 · doi:10.1509/jmkr.48.4.767
[38] Rossi P, Allenby G, McCulloch R (2005) Bayesian statistics in marketing. Wiley, New York · Zbl 1094.62037 · doi:10.1002/0470863692
[39] Sawtooth Software, Inc. (2013) The CBC system for choice-based conjoint analysis, Version 8, Sawtooth Software, Inc. http://www.sawtoothsoftware.com/download/techpap/cbctech.pdf. Accessed 28 Nov 2015
[40] Simonson I, Tversky A (1992) Choice in context: tradeoff contrasts and extremeness aversion. J Mark Res 29:281-295 · doi:10.2307/3172740
[41] Steiner WJ (2010) A Stackelberg-Nash model for new product design. OR Spectrum 32:21-48 · Zbl 1183.91027 · doi:10.1007/s00291-008-0137-4
[42] Steiner W, Hruschka H (2000) Conjointanalyse-basierte Produkt(linien)gestaltung unter Berücksichtigung von Konkurrenzreaktionen. OR Spektrum 22:71-95 · Zbl 0961.90029 · doi:10.1007/s002910050006
[43] Swait J (2003) Flexible covariance structures for categorical dependent variables through finite mixtures of generalized extreme value models. J Bus Econ Stat 21:80-87 · doi:10.1198/073500102288618784
[44] Train K (2003) Discrete choice methods with simulation. Cambridge University Press, Cambridge · Zbl 1047.62098 · doi:10.1017/CBO9780511753930
[45] Train K (2008) Em algorithms for nonparametric estimation of mixing distributions. J Choice Model 1:40-69 · doi:10.1016/S1755-5345(13)70022-8
[46] Tuma M, Decker R (2013) Finite mixture models in market segmentation: a review and suggestions for best practices. Electron J Bus Res Methods 11:2-15
[47] Vriens M, Wedel M, Wilms T (1996) Metric conjoint segmentation methods: a Monte Carlo comparison. J Mark Res 33:73-85 · doi:10.2307/3152014
[48] Wedel M, Kamakura W (2000) Market segmentation-conceptual and methodological foundations. Kluver Academic Publishers, Norwell · doi:10.1007/978-1-4615-4651-1
[49] Wedel M, Kamakura W, Arora N, Bemmaor NJC, Elrod T, Johnson R, Lenk P, Neslin S, Poulson C (1999) Discrete and continuous representations of unobserved heterogeneity in choice modeling. Mark Lett 10:219-232 · doi:10.1023/A:1008054316179
[50] Weeks M (1997) The multinomial probit model revisited: a discussion of parameter estimability, identification and specification testing. J Econ Surv 11:297-320 · doi:10.1111/1467-6419.00035
[51] Xu H, Craig B (2009) A probit latent class model with general correlation structures for evaluating accuracy of diagnostic tests. Biometrics 65:1145-1155 · Zbl 1180.62144 · doi:10.1111/j.1541-0420.2008.01194.x
[52] Yai T, Iwakura S, Morichi S (1997) Multinomial probit with structured covariance for route choice behavior. Transp Res B 31:195-207 · doi:10.1016/S0191-2615(96)00025-2
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.