Logarithmic law of large random correlation matrices

DOI10.3150/23-BEJ1600MaRDI QIDQ6178564zbMATH OpenFDO

Authors Nestor Parolya, Johannes Heiny, D. Kurowicka

Publication date 16 January 2024

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2103.13900

zbMATH Keywords

CLTrandom matrix theorydependent datasample correlation matrixlog determinantlarge-dimensional asymptotic

Mathematics Subject Classification ID

Multivariate analysis (62Hxx) Probability theory on algebraic and topological structures (60Bxx) Limit theorems in probability theory (60Fxx)

Abstract: Consider a random vector $mathbfy=mathbf{Sigma}^{1/2}mathbfx$, where the $p$ elements of the vector $mathbfx$ are i.i.d. real-valued random variables with zero mean and finite fourth moment, and $mathbf{Sigma}^{1/2}$ is a deterministic $pimesp$ matrix such that the spectral norm of the population correlation matrix $mathbfR$ of $mathbfy$ is uniformly bounded. In this paper, we find that the log determinant of the sample correlation matrix $hatmathbfR$ based on a sample of size $n$ from the distribution of $mathbfy$ satisfies a CLT (central limit theorem) for $p/nogammain(0,1]$ and $pleqn$. Explicit formulas for the asymptotic mean and variance are provided. In case the mean of $mathbfy$ is unknown, we show that after recentering by the empirical mean the obtained CLT holds with a shift in the asymptotic mean. This result is of independent interest in both large dimensional random matrix theory and high-dimensional statistical literature of large sample correlation matrices for non-normal data. At last, the obtained findings are applied for testing of uncorrelatedness of $p$ random variables. Surprisingly, in the null case $mathbfR=mathbfI$, the test statistic becomes completely pivotal and the extensive simulations show that the obtained CLT also holds if the moments of order four do not exist at all, which conjectures a promising and robust test statistic for heavy-tailed high-dimensional data.

Cites work

• scientific article; zbMATH DE number 3886886 (Why is no real title available?)
• scientific article; zbMATH DE number 5278585 (Why is no real title available?)
• scientific article; zbMATH DE number 1964693 (Why is no real title available?)
• A Dynamical Approach to Random Matrix Theory
• A central limit theorem for the determinant of a Wigner matrix
• A new test of independence for high-dimensional data
• Almost sure convergence of the largest and smallest eigenvalues of high-dimensional sample correlation matrices
• Asymptotic distribution of the largest off-diagonal entry of correlation matrices
• Asymptotic theory of statistics and probability
• CLT for linear spectral statistics of large-dimensional sample covariance matrices.
• Central limit theorems for classical likelihood ratio tests for high-dimensional normal distributions
• Concentration of measure and spectra of random matrices: applications to correlation matrices, elliptical distributions and beyond
• Determinant of sample correlation matrix with application
• Estimation of the global minimum variance portfolio in high dimensions
• Gaussian fluctuations for linear spectral statistics of large random covariance matrices
• Generating random correlation matrices based on partial correlations
• High dimensional correlation matrices: the central limit theorem and its applications
• Large sample correlation matrices: a comparison theorem and its applications
• Large sample covariance matrices and high-dimensional data analysis
• Likelihood ratio tests for high-dimensional normal distributions
• Likelihood ratio tests for many groups in high dimensions
• Likelihood ratio tests under model misspecification in high dimensions
• Limit theorems for random simplices in high dimensions
• Limiting laws of coherence of random matrices with applications to testing covariance structure and construction of compressed sensing matrices
• On Jiang's asymptotic distribution of the largest entry of a sample correlation matrix
• On moments of quadratic forms in non-spherically distributed variables
• On the sphericity test with large-dimensional observations
• Point process convergence for the off-diagonal entries of sample covariance matrices
• Random matrices: law of the determinant
• Spectral analysis of large dimensional random matrices
• Substitution principle for CLT of linear spectral statistics of high-dimensional sample covariance matrices with applications to hypothesis testing
• Testing for complete independence in high dimensions
• Testing for independence of large dimensional vectors
• The Distribution of the Determinant of a Complex Wishart Distributed Matrix
• The asymptotic distribution and Berry-Esseen bound of a new test for independence in high dimension with an application to stochastic optimization
• The asymptotic distribution of the determinant of a random correlation matrix
• The asymptotic distributions of the largest entries of sample correlation matrices.
• The distribution of volume reductions induced by isotropic random projections
• The limiting distributions of eigenvalues of sample correlation matrices
• The logarithmic law of random determinant
• The logarithmic law of sample covariance matrices near singularity
• Thin-shell theory for rotationally invariant random simplices
• Tracy-Widom law for the extreme eigenvalues of sample correlation matrices

Cited in

(2)

• Spectral statistics of sample block correlation matrices
• On the distribution of sample scale-free scatter matrices

This page was built for publication: Logarithmic law of large random correlation matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6178564)