Generalized Direct Sampling for Hierarchical Bayesian Models

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: 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