Two sample tests for high-dimensional covariance matrices

Abstract

We propose two tests for the equality of covariance matrices between two high-dimensional populations. One test is on the whole variance--covariance matrices, and the other is on off-diagonal sub-matrices, which define the covariance between two nonoverlapping segments of the high-dimensional random vectors. The tests are applicable (i) when the data dimension is much larger than the sample sizes, namely the "large $p$, small $n$" situations and (ii) without assuming parametric distributions for the two populations. These two aspects surpass the capability of the conventional likelihood ratio test. The proposed tests can be used to test on covariances associated with gene ontology terms.

Published in at http://dx.doi.org/10.1214/12-AOS993 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (13)

• Regularized estimation of large covariance matrices
• Covariance regularization by thresholding
• On testing the significance of sets of genes
• On Orthogonal and Symplectic Matrix Ensembles
• Higher criticism for detecting sparse heterogeneous mixtures
• A two-sample test for high-dimensional data with applications to gene-set testing
• The sparsity and bias of the Lasso selection in high-dimensional linear regression
• Random-set methods identify distinct aspects of the enrichment signal in gene-set analysis
• Corrections to LRT on Large Dimensional Covariance Matrix by RMT
• Tracy--Widom limit for the largest eigenvalue of a large class of complex sample covariance matrices
• Properties of higher criticism under strong dependence
• Multiple tests of association with biological annotation metadata
• To how many simultaneous hypothesis tests can normal, Student's t or bootstrap calibration be applied?

Cited by in corpus (50)

• Optimal hypothesis testing for high dimensional covariance matrices
• Deep Knockoffs
• Comparing large covariance matrices under weak conditions on the dependence structure and its application to gene clustering
• Change-point detection in high-dimensional covariance structure
• A high-dimensional two-sample test for the mean using random subspaces
• Inference for High-dimensional Differential Correlation Matrices
• A review of 20 years of naive tests of significance for high-dimensional mean vectors and covariance matrices
• Testing High Dimensional Covariance Matrices, with Application to Detecting Schizophrenia Risk Genes
• Bootstrapping High Dimensional Time Series
• A latent factor model with a mixture of sparse and dense factors to model gene expression data with confounding effects
• Extreme eigenvalues of large-dimensional spiked Fisher matrices with application
• Non-parametric shrinkage mean estimation for quadratic loss functions with unknown covariance matrices
• Sequential change-point detection in high-dimensional Gaussian graphical models
• Joint limiting laws for high-dimensional independence tests
• Two-Sample Testing in High-Dimensional Models
• Testing Hypotheses about Covariance Matrices in General MANOVA Designs
• Robust Multivariate Nonparametric Tests via Projection-Averaging
• Dynamic Linear Discriminant Analysis in High Dimensional Space
• Change Point Detection in the Mean of High-Dimensional Time Series Data under Dependence
• On Two Simple and Effective Procedures for High Dimensional Classification of General Populations
• Central Limit Theorems for Classical Likelihood Ratio Tests for High-Dimensional Normal Distributions
• Asymptotically Independent U-Statistics in High-Dimensional Testing
• Testing and Support Recovery of Correlation Structures for Matrix-Valued Observations with an Application to Stock Market Data
• Manifold Asymptotics of Quadratic-Form-Based Inference in Repeated Measures Designs
• Graph-Based Two-Sample Tests for Data with Repeated Observations
• Online Change-Point Detection in High-Dimensional Covariance Structure with Application to Dynamic Networks
• A Mixture Model to Detect Edges in Sparse Co-expression Graphs
• CLT for large dimensional general Fisher matrices and its applications in high-dimensional data analysis
• An Extreme-Value Approach for Testing the Equality of Large U-Statistic Based Correlation Matrices
• Substitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing
• Optimal modification of the LRT for the equality of two high-dimensional covariance matrices
• A Neighborhood-Assisted Hotelling's $T^2$ Test for High-Dimensional Means
• On High Dimensional Behaviour of Some Two-Sample Tests Based on Ball Divergence
• Online detection of local abrupt changes in high-dimensional Gaussian graphical models
• Multi-level Thresholding Test for High Dimensional Covariance Matrices
• Modified Pillai's trace statistics for two high-dimensional sample covariance matrices
• A self-organising eigenspace map for time series clustering
• Asymptotic power of likelihood ratio tests for high dimensional data
• High-dimensional Two-sample Precision Matrices Test: An Adaptive Approach through Multiplier Bootstrap
• Test of Covariance and Correlation Matrices
• Hypothesis Testing for Network Data with Power Enhancement
• Limit theorem associated with Wishart matrices with application to hypothesis testing for common principal components
• High Dimensional Tests for Functional Networks of Brain Anatomic Regions
• Compressed spectral screening for large-scale differential correlation analysis with application in selecting Glioblastoma gene modules
• Rank-adaptive covariance testing with applications to genomics and neuroimaging
• Optimal Structured Matrix Approximation for Robustness to Incomplete Biosequence Data
• Max-Type and Sum-Type Procedures for Online Change-Point Detection in the Mean of High-Dimensional Data
• Homogeneity Tests of Covariance and Change-Points Identification for High-Dimensional Functional Data
• High-dimensional tests for spherical location and spiked covariance
• A Global Homogeneity Test for High-Dimensional Linear Regression