Parallelizing MCMC via Weierstrass Sampler
Abstract
With the rapidly growing scales of statistical problems, subset based communication-free parallel MCMC methods are a promising future for large scale Bayesian analysis. In this article, we propose a new Weierstrass sampler for parallel MCMC based on independent subsets. The new sampler approximates the full data posterior samples via combining the posterior draws from independent subset MCMC chains, and thus enjoys a higher computational efficiency. We show that the approximation error for the Weierstrass sampler is bounded by some tuning parameters and provide suggestions for choice of the values. Simulation study shows the Weierstrass sampler is very competitive compared to other methods for combining MCMC chains generated for subsets, including averaging and kernel smoothing.
Cite
@article{arxiv.1312.4605,
title = {Parallelizing MCMC via Weierstrass Sampler},
author = {Xiangyu Wang and David B. Dunson},
journal= {arXiv preprint arXiv:1312.4605},
year = {2014}
}
Comments
The original Algorithm 1 removed. Provided some theoretical justification for refinement sampling (Theorem 2). Added a new algorithm in addition to the rejection sampling for handling dimensionality curse. New simulations and graphs (with new colors and designs). A real data analysis is also provided