|
Activity Number:
|
43
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 31, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
|
Abstract #320319
|
|
|
Title:
|
Consensus and Disagreement Among Distributions for Bayesian Inference
|
|
Author(s):
|
Ehsan Soofi* and Mehdi Shoja
|
|
Companies:
|
University of Wisconsin - Milwaukee and Citigroup
|
|
Keywords:
|
Entropy ;
Kullback-Leibler divergence ;
Mutual information ;
AEconometric models ;
Subset seclection
|
|
Abstract:
|
The consensus posterior or predictive distribution is defined by the mixtures of a set of posterior or predictive distributions calculated by Bayesian model averaging. The Jensen-Shannon divergence of the consensus distribution provides a measure of disagreement among the constituents of the mixture. Using this measure we identify posterior and predictive distributions that are represented well or not so well by the respective consensus distribution. We illustrate applications of this procedure to the posterior and predictive distributions of some well-known econometrics models.
|
Authors who are presenting talks have a * after their name.
