Online Program Home
My Program

Abstract Details

Activity Number: 470 - Recent Theoretical Advancements for MCMC Algorithms
Type: Topic Contributed
Date/Time: Wednesday, August 1, 2018 : 8:30 AM to 10:20 AM
Sponsor: Section on Bayesian Statistical Science
Abstract #329313 Presentation
Title: Selection of Proposal Distributions for Multiple Importance Sampling
Author(s): Vivekananda Roy* and Evangelos Evangelou
Companies: Iowa State University and University of Bath
Keywords: Bayes factor; marginal likelihood; polynomial ergodicity; reverse logistic estimator; sensitivity analysis

The standard importance sampling (IS) method uses samples from a single proposal distribution and assigns weights to them, according to the ratio of the target and proposal pdfs. This naive IS estimator, generally does not work well in multiple target examples as the weights can take arbitrarily large values making the estimator highly unstable. In such situations, alternative generalized IS estimators involving samples from multiple proposal distributions are preferred. Just like the standard IS, the success of these multiple IS estimators crucially depends on the choice of the proposal distributions. For selecting these proposal distributions, we propose three methods based on a nonparametric coverage criterion, the maximum entropy approach and a sequential approach, respectively. The proposed methods for selecting proposal densities are illustrated using several detailed examples.

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

Back to the full JSM 2018 program