Spectral measure of empirical autocovariance matrices of high-dimensional Gaussian stationary processes

DOI10.1142/S2010326322500538zbMATH Open1524.62412arXiv2110.08523WikidataQ114071573 ScholiaQ114071573MaRDI QIDQ6133488FDOQ6133488

Authors: Arup Bose, W. Hachem 

Publication date: 24 July 2023

Published in: Random Matrices: Theory and Applications (Search for Journal in Brave)

Abstract: Consider the empirical autocovariance matrix at a given non-zero time lag based on observations from a multivariate complex Gaussian stationary time series. The spectral analysis of these autocovariance matrices can be useful in certain statistical problems, such as those related to testing for white noise. We study the behavior of their spectral measures in the asymptotic regime where the time series dimension and the observation window length both grow to infinity, and at the same rate. Following a general framework in the field of the spectral analysis of large random non-Hermitian matrices, at first the probabilistic behavior of the small singular values of the shifted versions of the autocovariance matrix are obtained. This is then used to infer about the large sample behaviour of the empirical spectral measure of the autocovariance matrices at any lag. Matrix orthogonal polynomials on the unit circle play a crucial role in our study.

Full work available at URL: https://arxiv.org/abs/2110.08523

Recommendations

• Spectral analysis of sample autocovariance matrices of a class of linear time series in moderately high dimensions
• On the behavior of large empirical autocovariance matrices between the past and the future
• Large sample behaviour of high dimensional autocovariance matrices
• Limiting spectral distribution of sample autocovariance matrices
• Large covariance and autocovariance matrices

zbMATH Keywords

matrix orthogonal polynomialslimit spectral distributionmultivariate stationary processessmall singular valueslarge non-Hermitian matrix theoryhigh-dimensional times series analysis

Mathematics Subject Classification ID

Diagnostics, and linear inference and regression (62J20) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Random matrices (probabilistic aspects) (60B20) Random measures (60G57) Other special orthogonal polynomials and functions (33C47) Free probability and free operator algebras (46L54)

Cites Work

• Title not available (Why is that?)
• Title not available (Why is that?)
• Random matrices: universality of ESDs and the circular law
• High-dimensional probability. An introduction with applications in data science
• Title not available (Why is that?)
• Non-Hermitian random matrices with a variance profile. I: Deterministic equivalents and limiting esds
• Eigenvalue distribution of large random matrices
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Orthogonal polynomials
• Title not available (Why is that?)
• Title not available (Why is that?)
• Invertibility of symmetric random matrices
• Title not available (Why is that?)
• On the empirical distribution of eigenvalues of a class of large dimensional random matrices
• Large sample behaviour of high dimensional autocovariance matrices
• On Matrix-Valued Herglotz Functions
• On the least singular value of random symmetric matrices
• Time series. Data analysis and theory.
• Spectral Properties of Banded Toeplitz Matrices
• The Littlewood-Offord problem and invertibility of random matrices
• Smallest singular value of a random rectangular matrix
• Around the circular law
• Smallest singular value of random matrices and geometry of random polytopes
• Deterministic equivalents for certain functionals of large random matrices
• Small ball probabilities for linear images of high-dimensional distributions
• The circular law for random matrices
• The Analytic Theory of Matrix Orthogonal Polynomials
• A Simple Approach to the Global Regime of Gaussian Ensembles of Random Matrices
• Title not available (Why is that?)
• Inverses of Toeplitz Operators, Innovations, and Orthogonal Polynomials
• Non-hermitian random matrix theory: Method of hermitian reduction
• Necessary and sufficient condition that the limit of Stieltjes transforms is a Stieltjes transform
• On a general moment problem on the half axis
• A new method for bounding rates of convergence of empirical spectral distributions
• Joint convergence of sample autocovariance matrices when \(p/n\to 0\) with application
• Limits of infinite interaction radius, dimensionality and the number of components for random operators with off-diagonal randomness
• Smallest singular value and limit eigenvalue distribution of a class of non-Hermitian random matrices with statistical application
• Large covariance and autocovariance matrices
• Erratum: The limiting spectral distribution in terms of spectral density

Cited In (2)

• Spectral analysis of sample autocovariance matrices of a class of linear time series in moderately high dimensions
• Joint convergence of sample cross-covariance matrices

This page was built for publication: Spectral measure of empirical autocovariance matrices of high-dimensional Gaussian stationary processes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6133488)