Optimal hypothesis testing for high dimensional covariance matrices

Abstract

This paper considers testing a covariance matrix $\Sigma$ in the high dimensional setting where the dimension $p$ can be comparable or much larger than the sample size $n$. The problem of testing the hypothesis $H_0:\Sigma ={\Sigma }_0$ for a given covariance matrix ${\Sigma }_0$ is studied from a minimax point of view. We first characterize the boundary that separates the testable region from the non-testable region by the Frobenius norm when the ratio between the dimension $p$ over the sample size $n$ is bounded. A test based on a $U$-statistic is introduced and is shown to be rate optimal over this asymptotic regime. Furthermore, it is shown that the power of this test uniformly dominates that of the corrected likelihood ratio test (CLRT) over the entire asymptotic regime under which the CLRT is applicable. The power of the $U$-statistic based test is also analyzed when $p/n$ is unbounded.

Published in at http://dx.doi.org/10.3150/12-BEJ455 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)

References in corpus (4)

• Two sample tests for high-dimensional covariance matrices
• Corrections to LRT on Large Dimensional Covariance Matrix by RMT
• Asymptotic power of sphericity tests for high-dimensional data
• Simultaneous Inference of Covariances

Cited by in corpus (24)

• Optimal hypothesis testing for stochastic block models with growing degrees
• A review of 20 years of naive tests of significance for high-dimensional mean vectors and covariance matrices
• Sharp detection in PCA under correlations: all eigenvalues matter
• Optimal Covariance Change Point Localization in High Dimension
• Non-parametric shrinkage mean estimation for quadratic loss functions with unknown covariance matrices
• Joint limiting laws for high-dimensional independence tests
• Testing Hypotheses about Covariance Matrices in General MANOVA Designs
• Detecting Rare and Weak Spikes in Large Covariance Matrices
• Hypothesis testing for eigenspaces of covariance matrix
• Central Limit Theorems for Classical Likelihood Ratio Tests for High-Dimensional Normal Distributions
• Testing and Support Recovery of Correlation Structures for Matrix-Valued Observations with an Application to Stock Market Data
• Asymptotically Independent U-Statistics in High-Dimensional Testing
• Testing mutual independence in high dimension via distance covariance
• Testing for independence in high dimensions based on empirical copulas
• Regularized LRT for Large Scale Covariance Matrices: One Sample Problem
• The Sample Complexity of Robust Covariance Testing
• Adaptive test for large covariance matrices with missing observations
• Tests for Large Dimensional Covariance Structure Based on Rao's Score Test
• Asymptotic power of likelihood ratio tests for high dimensional data
• On the asymptotic distribution of the maximum sample spectral coherence of Gaussian time series in the high dimensional regime
• Asymptotic power of Rao's score test for independence in high dimensions
• High Dimensional Tests for Functional Networks of Brain Anatomic Regions
• Modifications of Wald's score tests on large dimensional covariance matrices structure
• StarTrek: Combinatorial Variable Selection with False Discovery Rate Control