Operator norm consistent estimation of large-dimensional sparse covariance matrices

Abstract

Estimating covariance matrices is a problem of fundamental importance in multivariate statistics. In practice it is increasingly frequent to work with data matrices $X$ of dimension $n\times p$, where $p$ and $n$ are both large. Results from random matrix theory show very clearly that in this setting, standard estimators like the sample covariance matrix perform in general very poorly. In this "large $n$, large $p$" setting, it is sometimes the case that practitioners are willing to assume that many elements of the population covariance matrix are equal to 0, and hence this matrix is sparse. We develop an estimator to handle this situation. The estimator is shown to be consistent in operator norm, when, for instance, we have $p\asymp n$ as $n\to \infty$. In other words the largest singular value of the difference between the estimator and the population covariance matrix goes to zero. This implies consistency of all the eigenvalues and consistency of eigenspaces associated to isolated eigenvalues. We also propose a notion of sparsity for matrices, that is, "compatible" with spectral analysis and is independent of the ordering of the variables.

Published in at http://dx.doi.org/10.1214/07-AOS559 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

References in corpus (3)

• Regularized estimation of large covariance matrices
• Covariance regularization by thresholding
• Tracy--Widom limit for the largest eigenvalue of a large class of complex sample covariance matrices

Cited by in corpus (45)

• Covariance regularization by thresholding
• Sparse permutation invariant covariance estimation
• Sparsistency and rates of convergence in large covariance matrix estimation
• Latent variable graphical model selection via convex optimization
• Optimal rates of convergence for covariance matrix estimation
• A Unified Framework for High-Dimensional Analysis of M-Estimators with Decomposable Regularizers
• Optimal detection of sparse principal components in high dimension
• Asymptotic normality and optimalities in estimation of large Gaussian graphical models
• Covariance Estimation: The GLM and Regularization Perspectives
• Optimal rates of convergence for sparse covariance matrix estimation
• Vast volatility matrix estimation for high-frequency financial data
• Adaptive covariance matrix estimation through block thresholding
• Covariance matrix estimation for stationary time series
• High-dimensional analysis of semidefinite relaxations for sparse principal components
• High-dimensionality effects in the Markowitz problem and other quadratic programs with linear constraints: Risk underestimation
• The spectrum of kernel random matrices
• High-dimensional variable selection via tilting
• Bayesian Sparse Factor Analysis of Genetic Covariance Matrices
• Asymptotic performance of PCA for high-dimensional heteroscedastic data
• Gaussian and bootstrap approximations for high-dimensional U-statistics and their applications
• Posterior contraction in sparse Bayesian factor models for massive covariance matrices
• Optimal sparse volatility matrix estimation for high-dimensional Itô processes with measurement errors
• Posterior convergence rates for estimating large precision matrices using graphical models
• Improved Sparse Low-Rank Matrix Estimation
• Regression on manifolds: Estimation of the exterior derivative
• A penalized empirical likelihood method in high dimensions
• On information plus noise kernel random matrices
• Convergence of the largest eigenvalue of normalized sample covariance matrices when p and n both tend to infinity with their ratio converging to zero
• Learning the Conditional Independence Structure of Stationary Time Series: A Multitask Learning Approach
• Robust Sparse Covariance Estimation by Thresholding Tyler's M-Estimator
• Introduction to correlation networks: Interdisciplinary approaches beyond thresholding
• Regularized estimation of linear functionals of precision matrices for high-dimensional time series
• Estimation of Large Covariance and Precision Matrices from Temporally Dependent Observations
• Detecting positive correlations in a multivariate sample
• On the Nyström and Column-Sampling Methods for the Approximate Principal Components Analysis of Large Data Sets
• Perturbations and projections of Kalman-Bucy semigroups
• Non-Asymptotic Analysis of Ensemble Kalman Updates: Effective Dimension and Localization
• Estimation of functionals of sparse covariance matrices
• SURE Information Criteria for Large Covariance Matrix Estimation and Their Asymptotic Properties
• Discussion of: Treelets--An adaptive multi-scale basis for sparse unordered data
• Regularizing stock return covariance matrices via multiple testing of correlations
• Nonasymptotic estimation and support recovery for high dimensional sparse covariance matrices
• Detecting the large entries of a sparse covariance matrix in sub-quadratic time
• Covariance Operator Estimation via Adaptive Thresholding
• Hub Detection in Gaussian Graphical Models