A Robust Bayesian Approach to Multinomial Choice Modeling

• Activity Number: 495
• Type: Contributed
• Date/Time: Wednesday, August 7, 2013 : 8:30 AM to 10:20 AM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Dries Benoit*+ and Stefan Van Aelst and Dirk Van den Poel
• Companies: Ghent University and Ghent University and Ghent University
• Keywords: robustness ; multinomial models ; choice modeling ; multivariate Laplace distribution

Abstract

A Bayesian method for robust estimation of multinomial choice models is presented. The method can be used for both correlated as well as uncorrelated choice alternatives. To account for outliers in the response direction, the fat tailed multivariate Laplace distribution is used. In addition, a shrinkage procedure is applied to handle outliers in the independent variables as well. By exploiting the scale mixture of normals representation of the multivariate Laplace distribution, an efficient Gibbs sampling algorithm is developed. A simulation study shows that estimation of the model parameters is less influenced by outliers compared to non-robust alternatives, even when the data generating model deviates from the assumed model. An analysis of margarine scanner data shows how our method can be used for better pricing decisions.