Generalized Direct Sampling for Hierarchical Bayesian Models

30 Pages Posted: 10 Aug 2011 Last revised: 26 Sep 2012

See all articles by Michael Braun

Michael Braun

Southern Methodist University (SMU) - Marketing Department

Paul Damien

University of Texas at Austin - McCombs School of Business

Date Written: August 7, 2012

Abstract

We develop a new method to sample from posterior distributions in hierarchical models without using Markov chain Monte Carlo. This method, which is a variant of importance 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. Additionally, the method can be used to compute marginal likelihoods.

Keywords: Bayesian Inference, Importance Sampling, Markov Chain Monte Carlo, Marginal Likelihood

JEL Classification: C11, C15, C63, C8, M3

Suggested Citation

Braun, Michael and Damien, Paul, Generalized Direct Sampling for Hierarchical Bayesian Models (August 7, 2012). McCombs Research Paper Series No. IROM-02-11; MIT Sloan Research Paper No. 4925-11. Available at SSRN: https://ssrn.com/abstract=1907835 or http://dx.doi.org/10.2139/ssrn.1907835

Michael Braun (Contact Author)

Southern Methodist University (SMU) - Marketing Department ( email )

United States

Paul Damien

University of Texas at Austin - McCombs School of Business ( email )

Austin, TX 78712
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
116
Abstract Views
970
rank
240,801
PlumX Metrics