Relative perturbation bounds with applications to empirical covariance operators

DOI10.1016/J.AIM.2022.108808zbMATH Open1505.60014arXiv1802.02869OpenAlexW2787201183MaRDI QIDQ2111217FDOQ2111217

Authors: Moritz Jirak, Martin Wahl 

Publication date: 28 December 2022

Published in: Advances in Mathematics (Search for Journal in Brave)

Abstract: The goal of this paper is to establish relative perturbation bounds, tailored for empirical covariance operators. Our main results are expansions for empirical eigenvalues and spectral projectors, leading to concentration inequalities and limit theorems. One of the key ingredients is a specific separation measure for population eigenvalues, which we call the relative rank, giving rise to a sharp invariance principle in terms of limit theorems, concentration inequalities and inconsistency results. Our framework is very general, requiring only $p>4$ moments and allows for a huge variety of dependence structures.

Full work available at URL: https://arxiv.org/abs/1802.02869

Recommendations

• Perturbation bounds for eigenspaces under a relative gap condition
• Asymptotics and concentration bounds for bilinear forms of spectral projectors of sample covariance
• Concentration inequalities and moment bounds for sample covariance operators
• Normal approximation and concentration of spectral projectors of sample covariance
• An \(\ell_{\infty}\) eigenvector perturbation bound and its application

zbMATH Keywords

consistencyprincipal component analysiscentral limit theoremperturbation theoryconcentration inequalitieslaw of large numberslimit theoremscovariance operatorrelative bounds

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Inequalities involving eigenvalues and eigenvectors (15A42) Random matrices (probabilistic aspects) (60B20) Central limit and other weak theorems (60F05) Perturbation theory of linear operators (47A55)

Cites Work

• Principal component analysis.
• Functional data analysis.
• Title not available (Why is that?)
• Title not available (Why is that?)
• On the distribution of the largest eigenvalue in principal components analysis
• Mathematical Foundations of Infinite-Dimensional Statistical Models
• Concentration inequalities. A nonasymptotic theory of independence
• Theoretical Foundations of Functional Data Analysis, with an Introduction to Linear Operators
• Asymptotics of sample eigenstructure for a large dimensional spiked covariance model
• Asymptotic theory for the principal component analysis of a vector random function: Some applications to statistical inference
• Weakly dependent functional data
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• Sharp nonasymptotic bounds on the norm of random matrices with independent entries
• Some estimates of norms of random matrices
• Title not available (Why is that?)
• User-friendly tail bounds for sums of random matrices
• Asymptotics of empirical eigenstructure for high dimensional spiked covariance
• Tracy-Widom limit for the largest eigenvalue of a large class of complex sample covariance matrices
• A Bernstein type inequality and moderate deviations for weakly dependent sequences
• Perturbation theory for linear operators.
• A bound on tail probabilities for quadratic forms in independent random variables whose distributions are not necessarily symmetric
• Phase transition of the largest eigenvalue for nonnull complex sample covariance matrices
• Asymptotic equivalence of functional linear regression and a white noise inverse problem
• Analytic perturbation theory for matrices and operators. Licensed ed
• Title not available (Why is that?)
• Theory for high-order bounds in functional principal components analysis
• Title not available (Why is that?)
• Methodology and convergence rates for functional linear regression
• Title not available (Why is that?)
• Random covariance matrices: universality of local statistics of eigenvalues
• Minimax and adaptive prediction for functional linear regression
• Large deviations of sums of independent random variables
• A note on the Hanson-Wright inequality for random vectors with dependencies
• New dependence coefficients. Examples and applications to statistics
• The eigenvalues and eigenvectors of finite, low rank perturbations of large random matrices
• CLT in functional linear regression models
• Random matrix approximation of spectra of integral operators
• Nonparametric estimation of scalar diffusions based on low frequency data
• Title not available (Why is that?)
• Optimal eigen expansions and uniform bounds
• Characterization of LIL behavior in Banach space
• Minimax adaptive tests for the functional linear model
• Perturbation Bounds for Matrix Eigenvalues
• Statistical properties of kernel principal component analysis
• Concentration inequalities and moment bounds for sample covariance operators
• Asymptotics and concentration bounds for bilinear forms of spectral projectors of sample covariance
• THE EIGEN- AND SINGULAR VALUES OF THE SUM AND PRODUCT OF LINEAR OPERATORS
• Optimal estimation and rank detection for sparse spiked covariance matrices
• Normal approximation and concentration of spectral projectors of sample covariance
• Singular vectors under random perturbation
• An overview of relative \(\sin\Theta\) theorems for invariant subspaces of complex matrices
• Nonasymptotic upper bounds for the reconstruction error of PCA
• The dimension-free structure of nonhomogeneous random matrices
• Random perturbation of low rank matrices: improving classical bounds
• Asymptotically efficient estimation of smooth functionals of covariance operators
• Upper and Lower Bounds for Stochastic Processes
• Second order accurate distributed eigenvector computation for extremely large matrices
• Limiting laws for divergent spiked eigenvalues and largest nonspiked eigenvalue of sample covariance matrices
• Perturbation bounds for eigenspaces under a relative gap condition

Cited In (6)

• Higher-order principal component analysis for the approximation of tensors in tree-based low-rank formats
• Mitigating the influence of the boundary on PDE-based covariance operators
• A note on the prediction error of principal component regression in high dimensions
• Quantitative limit theorems and bootstrap approximations for empirical spectral projectors
• Van Trees inequality, group equivariance, and estimation of principal subspaces
• Statistical inference for structured high-dimensional models. Abstracts from the workshop held March 11--17, 2018

Uses Software

• fda (R)

This page was built for publication: Relative perturbation bounds with applications to empirical covariance operators

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2111217)