Cross-Validated Loss-based Covariance Matrix Estimator Selection in High Dimensions

DOI10.1080/10618600.2022.2110883arXiv2102.09715OpenAlexW3131114373WikidataQ114099320 ScholiaQ114099320MaRDI QIDQ6094089FDOQ6094089

Authors: Nima S. Hejazi, Mark J. Van der Laan, Sandrine Dudoit 

Publication date: 9 October 2023

Published in: Journal of Computational and Graphical Statistics (Search for Journal in Brave)

Abstract: The covariance matrix plays a fundamental role in many modern exploratory and inferential statistical procedures, including dimensionality reduction, hypothesis testing, and regression. In low-dimensional regimes, where the number of observations far exceeds the number of variables, the optimality of the sample covariance matrix as an estimator of this parameter is well-established. High-dimensional regimes do not admit such a convenience, however. As such, a variety of estimators have been derived to overcome the shortcomings of the sample covariance matrix in these settings. Yet, the question of selecting an optimal estimator from among the plethora available remains largely unaddressed. Using the framework of cross-validated loss-based estimation, we develop the theoretical underpinnings of just such an estimator selection procedure. In particular, we propose a general class of loss functions for covariance matrix estimation and establish finite-sample risk bounds and conditions for the asymptotic optimality of the cross-validated estimator selector with respect to these loss functions. We evaluate our proposed approach via a comprehensive set of simulation experiments and demonstrate its practical benefits by application in the exploratory analysis of two single-cell transcriptome sequencing datasets. A free and open-source software implementation of the proposed methodology, the cvCovEst R package, is briefly introduced.

Full work available at URL: https://arxiv.org/abs/2102.09715

zbMATH Keywords

dimension reductioncovariance matrix estimationcross-validationhigh-dimensional statisticsloss-based estimation

Mathematics Subject Classification ID

Statistics (62-XX)

Cites Work

• The elements of statistical learning. Data mining, inference, and prediction
• Forecasting Using Principal Components From a Large Number of Predictors
• Inferential Theory for Factor Models of Large Dimensions
• High dimensional covariance matrix estimation using a factor model
• Title not available (Why is that?)
• Title not available (Why is that?)
• Large Covariance Estimation by Thresholding Principal Orthogonal Complements
• Projected principal component analysis in factor models
• Covariance regularization by thresholding
• Sparsistency and rates of convergence in large covariance matrix estimation
• A well-conditioned estimator for large-dimensional covariance matrices
• Nonlinear shrinkage estimation of large-dimensional covariance matrices
• On the distribution of the largest eigenvalue in principal components analysis
• Regularized estimation of large covariance matrices
• On consistency and sparsity for principal components analysis in high dimensions
• Asymptotics of the principal components estimator of large factor models with weakly influential factors
• DISTRIBUTION OF EIGENVALUES FOR SOME SETS OF RANDOM MATRICES
• Asymptotics of cross-validated risk estimation in estimator selection and performance assess\-ment
• A distribution-free theory of nonparametric regression
• Adaptive thresholding for sparse covariance matrix estimation
• Title not available (Why is that?)
• Generalized thresholding of large covariance matrices
• Optimal rates of convergence for covariance matrix estimation
• Probability Inequalities for the Sum of Independent Random Variables
• Spectrum estimation: a unified framework for covariance matrix estimation and PCA in large dimensions
• Oracle inequalities for multi-fold cross validation
• The Empirical Bayes Approach to Statistical Decision Problems
• Covariance, subspace, and intrinsic Crame/spl acute/r-Rao bounds
• Robust covariance estimation for approximate factor models
• Tuning-parameter selection in regularized estimations of large covariance matrices

Cited In (3)

• Cross-validation of covariance structures using the frobenius matrix distance as a discrepancy function
• Title not available (Why is that?)
• Ultrahigh dimensional precision matrix estimation via refitted cross validation

This page was built for publication: Cross-Validated Loss-based Covariance Matrix Estimator Selection in High Dimensions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6094089)