An Empirical Bayes Procedure for Improving Individual-Level Estimates and Predictions from Finite Mixtures of Multinomial Logit Models

Journal of Business & Economic Statistics; January 2004; 22, 1; p. 121

5 Pages Posted: 14 Feb 2014

See all articles by Wagner A. Kamakura

Wagner A. Kamakura

Rice University

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department; University of Groningen - Faculty of Economics and Business

Date Written: 2003

Abstract

Unobserved heterogeneity in random utility choice models can be dealt with by specifying either a multinomial or a normal distribution of the coefficients, leading to finite mixture logit and mixed logit models. Focusing on the former, we show that individual-level estimates and predictions of finite mixtures estimated by maximizing the likelihood function can be improved through integration over the estimation error of the hyperparameters, using an empirical Bayes approach. We investigate the conjecture that this approach is more robust against departures of the underlying assumptions of the finite mixture model in two Monte Carlo studies. We show that our approach improves the performance of the finite mixture model in representing individual-level parameters and producing hold-out forecasts. We illustrate with two examples that our approach may offer advantages in empirical applications involving the analysis or heterogeneous choice data.

Keywords: Finite mixtures,Multinomial logit, Random utility models

Suggested Citation

Kamakura, Wagner A. and Wedel, Michel, An Empirical Bayes Procedure for Improving Individual-Level Estimates and Predictions from Finite Mixtures of Multinomial Logit Models (2003). Journal of Business & Economic Statistics; January 2004; 22, 1; p. 121. Available at SSRN: https://ssrn.com/abstract=2394888

Wagner A. Kamakura (Contact Author)

Rice University ( email )

6100 South Main Street
P.O. Box 1892
Houston, TX 77005-1892
United States
(713) 348-6307 (Phone)

Michel Wedel

University of Maryland - Robert H. Smith School of Business, Marketing Department ( email )

College Park, MD 20742
United States
301.405.2162 (Phone)
301.405.0146 (Fax)

HOME PAGE: http://www.rhsmith.umd.edu/marketing/faculty/wedel.html

University of Groningen - Faculty of Economics and Business ( email )

Postbus 72
9700 AB Groningen
Netherlands

Register to save articles to
your library

Register

Paper statistics

Downloads
22
Abstract Views
304
PlumX Metrics