Ohio State nav bar

Seminar: James Johndrow

Statistics Seminar Series
February 11, 2016
All Day
209 W. Eighteenth Ave. (EA), Room 170

Title

Approximations of Markov Chains and Bayesian Inference

Speaker

James Johndrow, Duke University

Abstract

The Markov Chain Monte Carlo method is the dominant paradigm for posterior computation in Bayesian analysis. It has long been common to control computation time by making approximations to the Markov transition kernel. Comparatively little attention has been paid to convergence and estimation error in these approximating Markov Chains. We propose a framework for assessing when to use approximations in MCMC algorithms, and how much error in the transition kernel should be tolerated to obtain optimal estimation performance with respect to a specified discrepancy measure and computational budget. The results require only ergodicity of the exact kernel and control of the kernel approximation accuracy. The theoretical framework is applied to approximations based on random subsets of data, low-rank approximations of Gaussian processes, and a novel approximating Markov chain for discrete mixture models.
 
This is joint work with Jonathan Mattingly, Sayan Mukherjee and David Dunson.