Abstract:Testing large covariance matrices is of fundamental importance in statistical analysis with high-dimensional data. In the past decade, three types of test statistics have been studied in the literature: quadratic form statistics, maximum form statistics, and their weighted combination. It is known that quadratic form statistics would suffer from low power against sparse alternatives and maximum form statistics would suffer from low power against dense alternatives. The weighted combination methods were introduced to enhance the power of quadratic form statistics or maximum form statistics when the weights are appropriately chosen. In this article, we provide a new perspective to exploit the full potential of quadratic form statistics and maximum form statistics for testing high-dimensional covariance matrices. We propose a scale-invariant power-enhanced test based on Fisher's method to combine the p -values of quadratic form statistics and maximum form statistics. After carefully studying the asymptotic joint distribution of quadratic form statistics and maximum form statistics, we first prove that the proposed combination method retains the correct asymptotic size under the Gaussian assumption, and we also derive a new Lyapunov-type bound for the joint distribution and prove the correct asymptotic size of the proposed method without requiring the Gaussian assumption. Moreover, we show that the proposed method boosts the asymptotic power against more general alternatives. Finally, we demonstrate the finite-sample performance in simulation studies and a real application. Supplementary materials for this article are available online.

Tests for Covariance Structures with High-Dimensional Repeated Measurements

A High Dimensional Nonparametric Test for Proportional Covariance Matrices

Tests for High-Dimensional Generalized Linear Models under General Covariance Structure

Specification Tests for Covariance Structures in High-Dimensional Statistical Models

A Robust Covariance Testing Approach for High-Throughput Data

Two Sample Tests for High Dimensional Covariance Matrices

HYPOTHESIS TESTING ON LINEAR STRUCTURES OF HIGH DIMENSIONAL COVARIANCE MATRIX.

Tests for High-Dimensional Covariance Matrices Using Random Matrix Projection

Testing Super-Diagonal Structure in High Dimensional Covariance Matrices

Testing High Dimensional Covariance Matrices Via Posterior Bayes Factor.

Testing for patterns and structures in covariance and correlation matrices

Testing covariates in high-dimensional regression

Hypothesis Testing for High-Dimensional Covariance Matrices.

Testing Covariance Structure of Large-Dimensional Data Based on Wald'S Score Test

Mean Test for High-Dimensional Data Based on Covariance Matrix with Linear Structures

Joint Test for Homogeneity of High-Dimensional Means and Covariance Matrices Using Maximum-Type Statistics

Corrections to LRT on Large Dimensional Covariance Matrix by RMT

Set-based differential covariance testing for high-throughput data

Fisher's Combined Probability Test for High-Dimensional Covariance Matrices

Testing for covariance matrices in time-varying coefficient panel data models with fixed effects

A High-Dimensional Test for Multivariate Analysis of Variance under a Low-Dimensional Factor Structure