Department of Mathematics, Applied Mathematics and Statistics

Navigation + Search
Home / Abstracts / Colloquium-April 22, 2016

Colloquium-April 22, 2016

Posted on April 13, 2016

Friday, April 22, 2016 (3:00 p.m. in Yost 306)

Title: Exact MCMC Using Approximations and Bernoulli Factories

Speaker: Radu Herbei (Associate Professor of Statistics, The Ohio State University)

Hosted by Jenný Brynjarsdóttir

Abstract: With the ever increasing complexity of models  used in modern science, there is a need for new computing strategies. Classical MCMC algorithms (Metropolis-Hastings, Gibbs) have difficulty handling very high-dimensional state spaces and models where likelihood evaluation is impossible. In this work we study a collection of models for which the likelihood cannot be evaluated exactly; however, it can be estimated unbiasedly  in an efficient way via distributed computing.  Such models include, but are not limited to cases where the data are discrete noisy observations from a class of diffusion processes or partial measurements of a solution to a partial differential equation. In each case, an exact MCMC algorithm targeting the correct posterior distribution can be obtained either via the “auxiliary variable trick” or by using a Bernoulli factory to advance the current state.  We explore the advantages and disadvantages of such an MCMC algorithm and show how it can be used in an oceanographic application.

Page last modified: April 13, 2016