Scalable Rejection Sampling for Bayesian Hierarchical Models

• Activity Number: 403
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Marketing
• Author(s): Michael Braun*+ and Paul Damien
• Companies: SMU Cox School of Business and University of Texas at Austin
• Keywords: Bayesian methods ; Statistical computation ; Marketing research ; Customer analytics ; Heterogeneity ; Hierarchical models

Abstract

We develop a new method to sample from posterior distributions in Bayesian hierarchical models, as commonly used in marketing research, without using Markov chain Monte Carlo. This method, which is a variant of rejection sampling ideas, is generally applicable to high-dimensional models involving large data sets. Samples are independent, so they can be collected in parallel, and we do not need to be concerned with issues like chain convergence and autocorrelation. The method is scalable under the assumption that heterogeneous units are conditionally independent, and it can also be used to compute marginal likelihoods.