Teaching Bayesian Statistics to Marketing and Business Students

Abstract:     
We discuss our experiences teaching Bayesian Statistics to students in doctoral programs in business. These students often have weak backgrounds in mathematical statistics and a predisposition against likelihood-based methods stemming from prior exposure to econometrics. This can be overcome by an intense course which emphasizes the value of the Bayesian approach to solving non-trivial problems. The success of our course is primarily due to the emphasis on statistical computing. This is facilitated by our R package, bayesm, which provides efficient implementation of advanced methods and models.

Number of Pages in PDF File: 8

Keywords: bayesm package, hierarchical models, posterior inference, R software

JEL Classification: C11, M1, M3

working papers series