Posterior Averaging Information Criterion

TL;DR

The paper introduces a novel Bayesian model selection criterion, the posterior averaging information criterion, which assesses models based on their predictive performance over the entire posterior distribution, even with non-informative priors.

Contribution

It develops a new model selection method grounded in Kullback-Leibler divergence that corrects bias in posterior mean log-likelihood, applicable to models with degenerate priors.

Findings

Performs well in small samples in normal and binomial models

Provides a theoretically sound alternative to existing criteria

Applicable even when the true model is not in the candidate set

Abstract

We propose a new model selection method, the posterior averaging information criterion, for Bayesian model assessment from a predictive perspective. The theoretical foundation is built on the Kullback-Leibler divergence to quantify the similarity between the proposed candidate model and the underlying true model. From a Bayesian perspective, our method evaluates the candidate models over the entire posterior distribution in terms of predicting a future independent observation. Without assuming that the true distribution is contained in the candidate models, the new criterion is developed by correcting the asymptotic bias of the posterior mean of the log-likelihood against its expected log-likelihood. It can be generally applied even for Bayesian models with degenerate non-informative prior. The simulation in both normal and binomial settings demonstrates decent small sample performance.