Optimal rates of convergence for sparse covariance matrix estimation

DOI10.1214/12-AOS998zbMATH Open1373.62247arXiv1302.3030MaRDI QIDQ741791FDOQ741791

Authors: Harrison H. Zhou, T. Tony Cai 

Publication date: 15 September 2014

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: This paper considers estimation of sparse covariance matrices and establishes the optimal rate of convergence under a range of matrix operator norm and Bregman divergence losses. A major focus is on the derivation of a rate sharp minimax lower bound. The problem exhibits new features that are significantly different from those that occur in the conventional nonparametric function estimation problems. Standard techniques fail to yield good results, and new tools are thus needed. We first develop a lower bound technique that is particularly well suited for treating "two-directional" problems such as estimating sparse covariance matrices. The result can be viewed as a generalization of Le Cam's method in one direction and Assouad's Lemma in another. This lower bound technique is of independent interest and can be used for other matrix estimation problems. We then establish a rate sharp minimax lower bound for estimating sparse covariance matrices under the spectral norm by applying the general lower bound technique. A thresholding estimator is shown to attain the optimal rate of convergence under the spectral norm. The results are then extended to the general matrix $el{l}_w$ operator norms for $1lewleinfty$. In addition, we give a unified result on the minimax rate of convergence for sparse covariance matrix estimation under a class of Bregman divergence losses.

Full work available at URL: https://arxiv.org/abs/1302.3030

Recommendations

• Optimal rates of convergence for covariance matrix estimation
• Minimax optimal estimation of high-dimensional sparse covariance matrices with missing data
• Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
• Minimax estimation of large covariance matrices under \(\ell_1\)-norm
• Sparsistency and rates of convergence in large covariance matrix estimation

zbMATH Keywords

covariance matrix estimationthresholdingminimax lower boundoptimal rate of convergencespectral normBregman divergenceFrobenius normAssouad's lemmaLe Cam's method

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Estimation in multivariate analysis (62H12)

Cites Work

• Asymptotic Statistics
• Covariance regularization by thresholding
• Sparsistency and rates of convergence in large covariance matrix estimation
• Regularized estimation of large covariance matrices
• Adapting to unknown sparsity by controlling the false discovery rate
• Title not available (Why is that?)
• Introduction to nonparametric estimation
• Asymptotic methods in statistical decision theory
• Title not available (Why is that?)
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• Convergence of estimates under dimensionality restrictions
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Adaptive thresholding for sparse covariance matrix estimation
• Title not available (Why is that?)
• Generalized thresholding of large covariance matrices
• Optimal rates of convergence for sparse covariance matrix estimation
• Geometrizing rates of convergence. II
• Title not available (Why is that?)
• Optimal rates of convergence for covariance matrix estimation
• Title not available (Why is that?)
• Low-rank kernel learning with Bregman matrix divergences
• Matrix Nearness Problems with Bregman Divergences
• Title not available (Why is that?)

Cited In (96)

• Minimax optimal estimation of general bandable covariance matrices
• Goodness-of-fit testing-based selection for large-\(p\)-small-\(n\) problems: a two-stage ranking approach
• A shrinkage principle for heavy-tailed data: high-dimensional robust low-rank matrix recovery
• Sparse principal component analysis for high‐dimensional stationary time series
• Operator norm consistent estimation of large-dimensional sparse covariance matrices
• Optimal rates of convergence for covariance matrix estimation
• Data science, big data and statistics
• Large covariance estimation for compositional data via composition-adjusted thresholding
• Sparse estimation of high-dimensional correlation matrices
• Statistical and computational limits for sparse matrix detection
• Asymptotically minimax empirical Bayes estimation of a sparse normal mean vector
• Adaptive thresholding for sparse covariance matrix estimation
• Optimal large-scale quantum state tomography with Pauli measurements
• Gaussian and bootstrap approximations for high-dimensional U-statistics and their applications
• Large volatility matrix estimation with factor-based diffusion model for high-frequency financial data
• High-dimensional robust precision matrix estimation: cellwise corruption under \(\epsilon \)-contamination
• The finite sample properties of sparse M-estimators with pseudo-observations
• Minimax estimation of large covariance matrices under \(\ell_1\)-norm
• Optimal Bayesian minimax rates for unconstrained large covariance matrices
• Nonparametric eigenvalue-regularized precision or covariance matrix estimator
• An efficient ADMM algorithm for high dimensional precision matrix estimation via penalized quadratic loss
• Posterior contraction in sparse Bayesian factor models for massive covariance matrices
• Optimal estimation and rank detection for sparse spiked covariance matrices
• On the optimality of sliced inverse regression in high dimensions
• Covariance and precision matrix estimation for high-dimensional time series
• Sparse PCA: optimal rates and adaptive estimation
• The spectral norm of random inner-product kernel matrices
• Robust inference of risks of large portfolios
• Bernstein-von Mises theorems for functionals of the covariance matrix
• Minimax rate-optimal estimation of high-dimensional covariance matrices with incomplete data
• Tyler's and Maronna's M-estimators: non-asymptotic concentration results
• Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation
• Optimal sparse volatility matrix estimation for high-dimensional Itô processes with measurement errors
• Sparse covariance matrix estimation in high-dimensional deconvolution
• Adaptive covariance matrix estimation through block thresholding
• Estimating large precision matrices via modified Cholesky decomposition
• Linear shrinkage estimation of large covariance matrices using factor models
• Estimation of matrices with row sparsity
• Bayesian inference for high-dimensional decomposable graphs
• High-dimensional autocovariance matrices and optimal linear prediction
• Minimax posterior convergence rates and model selection consistency in high-dimensional DAG models based on sparse Cholesky factors
• Sparse and low-rank covariance matrix estimation
• Estimation and Calibration of Lévy Models via Fourier Methods
• Large covariance estimation through elliptical factor models
• Inference for high-dimensional differential correlation matrices
• Asymptotic normality and optimalities in estimation of large Gaussian graphical models
• Weak convergence of the empirical spectral distribution of high-dimensional band sample covariance matrices
• Positive-definite modification of a covariance matrix by minimizing the matrix \(\ell_{\infty}\) norm with applications to portfolio optimization
• Sparse PCA-based on high-dimensional Itô processes with measurement errors
• Fixed support positive-definite modification of covariance matrix estimators via linear shrinkage
• Large Covariance Estimation by Thresholding Principal Orthogonal Complements
• Sparse Generalized Eigenvalue Problem: Optimal Statistical Rates via Truncated Rayleigh Flow
• Obtaining minimax lower bounds: a review
• Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas
• Rates of convergence in conditional covariance matrix with nonparametric entries estimation
• Rate-optimal posterior contraction for sparse PCA
• Law of log determinant of sample covariance matrix and optimal estimation of differential entropy for high-dimensional Gaussian distributions
• Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
• Robust covariance and scatter matrix estimation under Huber's contamination model
• Optimal computational and statistical rates of convergence for sparse nonconvex learning problems
• Optimal rates of convergence for sparse covariance matrix estimation
• Minimax optimal estimation of high-dimensional sparse covariance matrices with missing data
• Estimation and inference for precision matrices of nonstationary time series
• Sparse basis covariance matrix estimation for high dimensional compositional data via hard thresholding
• Inferences in panel data with interactive effects using large covariance matrices
• Inference for high‐dimensional linear models with locally stationary error processes
• Optimal covariance matrix estimation for high-dimensional noise in high-frequency data
• Nonparametric matrix regression function estimation over symmetric positive definite matrices
• Gaussian Patch Mixture Model Guided Low-Rank Covariance Matrix Minimization for Image Denoising
• Minimax estimation of covariance and precision matrices for high-dimensional time series with long-memory
• Estimation of a multiplicative correlation structure in the large dimensional case
• Minimax estimation of large precision matrices with bandable Cholesky factor
• Title not available (Why is that?)
• Bayesian joint inference for multiple directed acyclic graphs
• Affine invariant integrated rank-weighted statistical depth: properties and finite sample analysis
• A factor-GARCH model for high dimensional volatilities
• Rates of estimation for high-dimensional multireference alignment
• Robust Shape Matrix Estimation for High-Dimensional Compositional Data with Application to Microbial Inter-Taxa Analysis
• Rate-optimal robust estimation of high-dimensional vector autoregressive models
• Differentially private high dimensional sparse covariance matrix estimation
• Rank-based correlation matrix estimation for high dimensional microbiome data
• Lower bound estimation for a family of high-dimensional sparse covariance matrices
• A comparison of methods for estimating the determinant of high-dimensional covariance matrix
• Matrix means and a novel high-dimensional shrinkage phenomenon
• Post-processed posteriors for banded covariances
• Robust sparse covariance estimation by thresholding Tyler's M-estimator
• Testing and support recovery of correlation structures for matrix-valued observations with an application to stock market data
• Trimmed estimators for large dimensional sparse covariance matrices
• Exponent of cross-sectional dependence for residuals
• A Bernstein-type inequality for high dimensional linear processes with applications to robust estimation of time series regressions
• The beta-mixture shrinkage prior for sparse covariances with near-minimax posterior convergence rate
• Tight lower bound of sparse covariance matrix estimation in the local differential privacy model
• Bayesian sparse spiked covariance model with a continuous matrix shrinkage prior
• A Bayesian approach for partial Gaussian graphical models with sparsity
• Adaptive thresholding estimator for differential association structures in two independent contingency tables
• Post-processed posteriors for sparse covariances

This page was built for publication: Optimal rates of convergence for sparse covariance matrix estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q741791)