James O. Berger

Overview

James O. (Jim) Berger is a statistician (Arts and Sciences Professor of Statistics, Duke University; former SAMSI director) and one of the central figures in modern Bayesian statistics — statistical decision theory, objective/reference priors, Bayesian model selection, and multiple-testing.

Key Contributions / Features

• Objective Bayes and model selection: reference priors (Berger-Bernardo), intrinsic and fractional Bayes factors (Berger-Pericchi), and default priors for hypothesis testing; author of Statistical Decision Theory and Bayesian Analysis.
• Mixtures of g-priors (Liang et al. 2008): coauthor of the hyper-g prior families for Bayesian variable selection (Zellner's g-Prior).

Related

• Zellner's g-Prior
• Bayes Factor
• Bayesian Model Selection
• Liang et al. (2008) Mixtures of g-Priors for Bayesian Variable Selection
• Feng Liang
• Merlise Clyde