Asymptotics for high-dimensional covariance matrices and quadratic forms with applications to the trace functional and shrinkage

DOI10.1016/J.SPA.2017.10.007zbMATH Open1391.60061arXiv1711.01835OpenAlexW2601695767MaRDI QIDQ1639677FDOQ1639677

Authors: Ansgar Steland, Rainer von Sachs 

Publication date: 13 June 2018

Published in: Stochastic Processes and their Applications (Search for Journal in Brave)

Abstract: We establish large sample approximations for an arbitray number of bilinear forms of the sample variance-covariance matrix of a high-dimensional vector time series using $el{l}_1$-bounded and small $el{l}_2$-bounded weighting vectors. Estimation of the asymptotic covariance structure is also discussed. The results hold true without any constraint on the dimension, the number of forms and the sample size or their ratios. Concrete and potential applications are widespread and cover high-dimensional data science problems such as tests for large numbers of covariances, sparse portfolio optimization and projections onto sparse principal components or more general spanning sets as frequently considered, e.g. in classification and dictionary learning. As two specific applications of our results, we study in greater detail the asymptotics of the trace functional and shrinkage estimation of covariance matrices. In shrinkage estimation, it turns out that the asymptotics differs for weighting vectors bounded away from orthogonaliy and nearly orthogonal ones in the sense that their inner product converges to 0.

Full work available at URL: https://arxiv.org/abs/1711.01835

Recommendations

• Large-sample approximations for variance-covariance matrices of high-dimensional time series
• Shrinkage for covariance estimation: asymptotics, confidence intervals, bounds and applications in sensor monitoring and finance
• Matrix shrinkage of high-dimensional expectation vectors
• Nonlinear shrinkage estimation of large-dimensional covariance matrices
• Sample covariance shrinkage for high dimensional dependent data

zbMATH Keywords

Brownian motionlong memoryquadratic formstrong approximationtracelinear process

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Functional limit theorems; invariance principles (60F17)

Cites Work

• Testing significance of features by lassoed principal components
• A penalized matrix decomposition, with applications to sparse principal components and canonical correlation analysis
• High dimensional covariance matrix estimation using a factor model
• Asymptotics of eigenvalues and unit-length eigenvectors of sample variance and correlation matrices
• A well-conditioned estimator for large-dimensional covariance matrices
• Linear processes in function spaces. Theory and applications
• Title not available (Why is that?)
• Approximation and learning by greedy algorithms
• Strong invariance principles for dependent random variables
• Strong approximation for a class of stationary processes
• Sparse and stable Markowitz portfolios
• Covariances estimation for long-memory processes
• Almost sure invariance principles for weakly dependent vector-valued random variables
• On linear processes with dependent innovations
• Change-point analysis in increasing dimension
• A note on the almost sure approximation of weakly dependent random variables
• Strong approximations of martingale vectors and their applications in Markov-chain adaptive designs
• Strong approximation for cross-covariances of linear variables with long-range dependence
• Corrigendum to: ``Asymptotics of eigenvalues and unit-length eigenvectors of sample variance and correlation matrices
• Large-sample approximations for variance-covariance matrices of high-dimensional time series
• Sample covariance shrinkage for high dimensional dependent data

Cited In (5)

• Are minimum variance portfolios in multi-factor models long in low-beta assets?
• Detecting changes in the second moment structure of high-dimensional sensor-type data in a K-sample setting
• Testing and estimating change-points in the covariance matrix of a high-dimensional time series
• Flexible nonlinear inference and change-point testing of high-dimensional spectral density matrices
• Sample covariance shrinkage for high dimensional dependent data

Uses Software

• lpc

This page was built for publication: Asymptotics for high-dimensional covariance matrices and quadratic forms with applications to the trace functional and shrinkage

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1639677)