Online Program Home
My Program

Abstract Details

Activity Number: 347 - Computationally Intensive Bayesian Methodology
Type: Contributed
Date/Time: Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #305257
Title: Bayesian Sampling in Constrained Domains
Author(s): Sharang Chaudhry* and Kaushik Ghosh and Daniel Lautzenheiser
Companies: University of Nevada Las Vegas and University of Nevada Las Vegas and University of Nevada Las Vegas
Keywords: MCMC; Metropolis-Hastings; sum-to-one; simplex; domain constraint

Picking optimal proposal distributions for Bayesian sampling is a well known yet challenging problem. Its difficulty can get exacerbated when working with parameters that have domain constraints. In this work, a transformation called inversion in a sphere is used within the Metropolis-Hastings framework to make the constrained sets more amenable to sampling. The proposed scheme is demonstrated across two examples with comparative analyses and applications.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program