Statistics Seminar

James HobertUniversity of Florida
Convergence analysis of MCMC algorithms for Bayesian robust multivariate regression

Wednesday, November 8, 2017 - 4:15pm
Biotech G01

Let $\pi$ denote the intractable posterior density that results when the likelihood from a multivariate linear regression model with errors from a scale mixture of normals is combined with the standard non-informative prior. There is a simple data augmentation (DA) algorithm and a corresponding Haar PX-DA algorithm that can be used to explore $\pi$. I will explain how the behavior of the mixing density near the origin is related to the rate at which the corresponding Markov chains converge. (This is joint work with Yeun-Ji Jung, Kshitij Khare and Qian Qin.)