Sample Covariance Matrices of Heavy-Tailed Distributions

DOI10.1093/IMRN/RNX067MaRDI QIDQ4619459zbMATH OpenOpenAlexFDO

Authors Konstantin E. Tikhomirov

Publication date 6 February 2019

Published in IMRN. International Mathematics Research Notices (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1606.03557

zbMATH Keywords

covariance matrixheavy-tailed distribution

Mathematics Subject Classification ID

Random matrices (probabilistic aspects) (60B20) Large deviations (60F10) Statistics of extreme values; tail inference (62G32) Estimation in multivariate analysis (62H12)

Abstract: Let $p>2$, $Bgeq1$, $Ngeqn$ and let $X$ be a centered $n$-dimensional random vector with the identity covariance matrix such that $suplimit{s}_{ain{S}^{n-1}}mathrmE|langleX,aangle{|}^{p}leqB$. Further, let $X_1,X_2,dots,X_N$ be independent copies of $X$, and $Sigm{a}_N:=frac1Nsu{m}_{i=1}^N{X}_i{X_i}^T$ be the sample covariance matrix. We prove that ${\displaystyle }$K^{-1}|Sigma_N-I_n|_{2 o 2}leqfrac{1}{N}maxlimits_{ileq N}|X_i|^2 +Bigl(frac{n}{N}Bigr)^{1-2/p}log^4frac{N}{n}+Bigl(frac{n}{N}Bigr)^{1-2/min(p,4)}${\displaystyle }$ with probability at least $1-frac1n$, where $K>0$ depends only on $B$ and $p$. In particular, for all $p>4$ we obtain a quantitative Bai-Yin type theorem.

Recommendations

• Sample covariance matrix for random vectors with heavy tails
• Covariance Matrix Estimation With Heterogeneous Samples
• Large sample autocovariance matrices of linear processes with heavy tails
• User-friendly covariance estimation for heavy-tailed distributions
• scientific article; zbMATH DE number 4182499
• Large sample covariance matrices of Gaussian observations with uniform correlation decay
• Large sample covariance matrices and high-dimensional data analysis
• Singular vector distribution of sample covariance matrices
• Asymptotic theory for the sample covariance matrix of a heavy-tailed multivariate time series

Cited in

(33)

• Sample covariance matrix for random vectors with heavy tails
• Sub-Gaussian estimators of the mean of a random vector
• The smallest singular value of a shifted d-regular random square matrix
• Moment estimator for random vectors with heavy tails
• Marcinkiewicz-type discretization of L^p-norms under the Nikolskii-type inequality assumption
• On Monte-Carlo methods in convex stochastic optimization
• Exact minimax risk for linear least squares, and the lower tail of sample covariance matrices
• Random embeddings with an almost Gaussian distortion
• An isomorphic Dvoretzky-Milman theorem using general random ensembles
• Sampling discretization and related problems
• The lower tail of random quadratic forms with applications to ordinary least squares
• Polynomial threshold functions, hyperplane arrangements, and random tensors
• Approximating the covariance ellipsoid
• Extreme singular values of inhomogeneous sparse random rectangular matrices
• Large-dimensional central limit theorem with fourth-moment error bounds on convex sets and balls
• Covariance estimation under missing observations and L₄ - L₂ moment equivalence
• Asymptotic geometric analysis: achievements and perspective
• Improved covariance estimation: optimal robustness and sub-Gaussian guarantees under heavy tails
• Empirical approximation of the Gaussian distribution in \(\mathbb{R}^d\)
• Covariance estimation with direction dependence accuracy
• Covariance estimation: optimal dimension-free guarantees for adversarial corruption and heavy tails
• A geometrical analysis of kernel ridge regression and its applications
• Sharp concentration of simple random tensors
• A geometrical viewpoint on the benign overfitting property of the minimum _2-norm interpolant estimator and its universality
• Sharper dimension-free bounds on the Frobenius distance between sample covariance and its expectation
• Structure preservation via the Wasserstein distance
• How close is the sample covariance matrix to the actual covariance matrix?
• An upper bound on the smallest singular value of a square random matrix
• Sharp bounds for max-sliced Wasserstein distances
• Concentration and moment inequalities for sums of independent heavy-tailed random matrices
• Estimating eigenvectors and eigenspaces of covariance matrices: Optimal Bounds and Conditions for Consistency
• Comparison theorems for the minimum eigenvalue of a random positive-semidefinite matrix
• Moment bounds for large autocovariance matrices under dependence

This page was built for publication: Sample Covariance Matrices of Heavy-Tailed Distributions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4619459)