Outlier‐Robust Bayesian Multinomial Choice Modeling

摘要: Summary A Bayesian method for outlier-robust estimation of multinomial choice models is presented. The can be used both correlated as well uncorrelated alternatives and guarantees robustness towards outliers in the dependent independent variables. To account response direction, fat-tailed multivariate Laplace distribution used. Leverage points are handled via a shrinkage procedure. A simulation study shows that model parameters less influenced by compared to non-robust alternatives. An analysis margarine scanner data how our better pricing decisions. Copyright © 2015 John Wiley & Sons, Ltd.