Consistency of Bayes factor for nonnested model selection when the model dimension grows

TL;DR

This paper investigates the consistency of Bayes factors for nonnested linear model selection as the number of parameters increases, extending previous results from nested models and comparing different Bayes factor approaches.

Contribution

It generalizes Bayes factor consistency results to nonnested models with growing dimensions and compares these with intrinsic Bayes factors.

Findings

Bayes factor remains consistent under certain growth conditions.

Comparison shows similarities and differences between proposed and intrinsic Bayes factors.

Results extend existing theory from nested to nonnested models.

Abstract

Zellner's $g$-prior is a popular prior choice for the model selection problems in the context of normal regression models. Wang and Sun [J. Statist. Plann. Inference 147 (2014) 95-105] recently adopt this prior and put a special hyper-prior for $g$, which results in a closed-form expression of Bayes factor for nested linear model comparisons. They have shown that under very general conditions, the Bayes factor is consistent when two competing models are of order $O(n^{\tau})$ for $\tau <1$ and for $\tau=1$ is almost consistent except a small inconsistency region around the null hypothesis. In this paper, we study Bayes factor consistency for nonnested linear models with a growing number of parameters. Some of the proposed results generalize the ones of the Bayes factor for the case of nested linear models. Specifically, we compare the asymptotic behaviors between the proposed Bayes factor and the intrinsic Bayes factor in the literature.