Maximum Pairwise Bayes Factors for Covariance Structure Testing

TL;DR

This paper introduces a new Bayesian testing framework using maximum pairwise Bayes factors for covariance matrix structure testing, especially effective in high-dimensional and sparse settings, with proven optimality and practical scalability.

Contribution

It develops a novel, computationally scalable Bayesian test for large covariance matrices that is optimal against sparse alternatives and includes a graph selection method.

Findings

The proposed test is asymptotically optimal in distinguishing null and alternative hypotheses.

The covariance graph selection procedure is consistent and scalable.

Simulation studies show advantages over existing methods.

Abstract

Hypothesis testing of structure in covariance matrices is of significant importance, but faces great challenges in high-dimensional settings. Although consistent frequentist one-sample covariance tests have been proposed, there is a lack of simple, computationally scalable, and theoretically sound Bayesian testing methods for large covariance matrices. Motivated by this gap and by the need for tests that are powerful against sparse alternatives, we propose a novel testing framework based on the maximum pairwise Bayes factor. Our initial focus is on one-sample covariance testing; the proposed test can {\it optimally} distinguish null and alternative hypotheses in a frequentist asymptotic sense. We then propose diagonal tests and a scalable covariance graph selection procedure that are shown to be consistent. A simulation study evaluates the proposed approach relative to competitors. We illustrate advantages of our graph selection method on a gene expression data set.