Properties of linear spectral statistics of frequency-smoothed estimated spectral coherence matrix of high-dimensional Gaussian time series

DOI10.1214/21-EJS1923zbMATH Open1493.62348arXiv2111.08047OpenAlexW3214203227MaRDI QIDQ2074296FDOQ2074296

Authors: Philippe Loubaton, Alexis Rosuel 

Publication date: 9 February 2022

Published in: Electronic Journal of Statistics (Search for Journal in Brave)

Abstract: The asymptotic behaviour of Linear Spectral Statistics (LSS) of the smoothed periodogram estimator of the spectral coherency matrix of a complex Gaussian high-dimensional time series $(y_n{)}_{ninmathbbZ}$ with independent components is studied under the asymptotic regime where the sample size $N$ converges towards $+infty$ while the dimension $M$ of $y$ and the smoothing span of the estimator grow to infinity at the same rate in such a way that $fracMNightarrow0$. It is established that, at each frequency, the estimated spectral coherency matrix is close from the sample covariance matrix of an independent identically $mathcal{N}_{mathbbC}(0,I_M)$ distributed sequence, and that its empirical eigenvalue distribution converges towards the Marcenko-Pastur distribution. This allows to conclude that each LSS has a deterministic behaviour that can be evaluated explicitly. Using concentration inequalities, it is shown that the order of magnitude of the supremum over the frequencies of the deviation of each LSS from its deterministic approximation is of the order of $frac1M+fracsqrtMN+(fracMN{)}^3$ where $N$ is the sample size. Numerical simulations supports our results.

Full work available at URL: https://arxiv.org/abs/2111.08047

Recommendations

• On the asymptotic distribution of the maximum sample spectral coherence of Gaussian time series in the high dimensional regime
• Spectral analysis of sample autocovariance matrices of a class of linear time series in moderately high dimensions
• Large sample behaviour of high dimensional autocovariance matrices
• On the asymptotic behavior of the eigenvalue distribution of block correlation matrices of high-dimensional time series
• On the Marčenko-Pastur law for linear time series

zbMATH Keywords

time seriesindependence testspectral analysishigh dimensional statisticsrandom matrices

Mathematics Subject Classification ID

Hypothesis testing in multivariate analysis (62H15) Random matrices (probabilistic aspects) (60B20) Inference from stochastic processes and spectral analysis (62M15)

Cites Work

• Spectral analysis of large dimensional random matrices
• The asymptotic distributions of the largest entries of sample correlation matrices.
• Hanson-Wright inequality and sub-Gaussian concentration
• Nonparametric approach for non-Gaussian vector stationary processes
• Testing nonparametric and semiparametric hypotheses in vector stationary processes
• Eigenvalue distribution of large random matrices
• A Method for Testing the Independence of Two Time Series That Accounts for a Potential Pattern in the Cross-Correlation Function
• Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions
• The largest eigenvalues of finite rank deformation of large Wigner matrices: Convergence and nonuniversality of the fluctuations
• Averaging fluctuations in resolvents of random band matrices
• Deterministic equivalents for certain functionals of large random matrices
• Testing for independence between two covariance stationary time series
• Title not available (Why is that?)
• Checking the Independence of Two Covariance-Stationary Time Series: A Univariate Residual Cross-Correlation Approach
• Local semicircle law for Wigner matrices
• Title not available (Why is that?)
• The limiting distributions of eigenvalues of sample correlation matrices
• Testing independence with high-dimensional correlated samples
• Tests for noncorrelation of two multivariate ARMA time series
• Testing independence among a large number of high-dimensional random vectors
• A New Approach for Mutual Information Analysis of Large Dimensional Multi-Antenna Channels
• Title not available (Why is that?)
• Shrinkage estimation in the frequency domain of multivariate time series
• Random matrices with complex Gaussian entries
• On the almost sure location of the singular values of certain Gaussian block-Hankel large random matrices
• A frequency-domain based test for non-correlation between stationary time series
• Robust residual cross correlation tests for lagged relations in time series
• Title not available (Why is that?)
• Likelihood ratio tests for many groups in high dimensions
• Correlation Tests and Linear Spectral Statistics of the Sample Correlation Matrix

Cited In (4)

• Likelihood ratio tests under model misspecification in high dimensions
• On the asymptotic distribution of the maximum sample spectral coherence of Gaussian time series in the high dimensional regime
• Spectral analysis of sample autocovariance matrices of a class of linear time series in moderately high dimensions
• Spectral analysis of high-dimensional time series

This page was built for publication: Properties of linear spectral statistics of frequency-smoothed estimated spectral coherence matrix of high-dimensional Gaussian time series

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2074296)