Jim Hobert

Title: Asymptotically Stable Drift and Minorization for Markov Chains with Application to Albert and Chib's Algorithm

Abstract: Recent work suggests that, while often useful for establishing convergence rates when $n$ and $p$ are fixed, techniques based on drift and minorization conditions may not be versatile enough to handle the more delicate task of convergence complexity analysis.  In this talk, general methods of sharpening drift and minorization conditions are introduced, and an application of these ideas to Albert and Chib's (1995, JASA) data augmentation algorithm for the Bayesian probit model is described.  (This is joint work with Qian Qin.)