|
Activity Number:
|
302
- Advances in Bayesian Computation
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #305066
|
Presentation
|
|
Title:
|
A New Visualization for MCMC Output Analysis
|
|
Author(s):
|
Nathan Robertson* and James Flegal
|
|
Companies:
|
University of California, Riverside and University of California, Riverside
|
|
Keywords:
|
MCMC;
Quantiles;
Visualization;
Multivariate
|
|
Abstract:
|
In MCMC sampling one estimates features of a target distribution by generating dependent samples from an approximate distribution. Assessing the quality of estimation from these procedures has been studied for features that are expressible as expectations and quantiles. These settings cover posterior means and credible intervals as are commonly of interest in Bayesian analyses, however these procedures have typically been limited to univariate and marginal settings except in the case of posterior means. We develop a joint estimation procedure which is applicable to combinations of expectations and quantiles and provide a novel method which motivates a class of visualizations addressing the quality of estimation.
|
Authors who are presenting talks have a * after their name.
