Edge statistics of large dimensional deformed rectangular matrices

DOI10.1016/J.JMVA.2022.105051MaRDI QIDQ2079603zbMATH OpenOpenAlexWikidataFDO

Authors Xiucai Ding, Fan Yang

Publication date 30 September 2022

Published in Journal of Multivariate Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2009.00389

zbMATH Keywords

eigenvalue rigidityeigenvector delocalizationspectral statisticsrectangular free convolutionedge eigenvalueedge statisticsinterference-plus-noise matrix

Mathematics Subject Classification ID

Factor analysis and principal components; correspondence analysis (62H25) Eigenvalues, singular values, and eigenvectors (15A18) Random matrices (algebraic aspects) (15B52) Random matrices (probabilistic aspects) (60B20) Free probability and free operator algebras (46L54)

Abstract: We consider the edge statistics of large dimensional deformed rectangular matrices of the form $Y_t=Y+sqrttX,$ where $Y$ is a $pimesn$ deterministic signal matrix whose rank is comparable to $n$, $X$ is a $pimesn$ random noise matrix with centered i.i.d. entries with variance $n^{-1}$, and $t>0$ gives the noise level. This model is referred to as the interference-plus-noise matrix in the study of massive multiple-input multiple-output (MIMO) system, which belongs to the category of the so-called signal-plus-noise model. For the case $t=1$, the spectral statistics of this model have been studied to a certain extent in the literature. In this paper, we study the singular value and singular vector statistics of $Y_t$ around the right-most edge of the singular value spectrum in the harder regime $n^{-2/3}lltll1$. This regime is harder than the $t=1$ case, because on one hand, the edge behavior of the empirical spectral distribution (ESD) of $Y{Y}^{o}p$ has a strong effect on the edge statistics of $Y_t{Y}_t^{o}p$ since $tll1$ is "small", while on the other hand, the edge statistics of $Y_t$ is also not merely a perturbation of those of $Y$ since $tgg{n}^{-2/3}$ is "large". Under certain regularity assumptions on $Y,$ we prove the edge universality, eigenvalues rigidity and eigenvector delocalization for the matrices $Y_t{Y}_t^{o}p$ and $Y_t^{o}p{Y}_t$. These results can be used to estimate and infer the massive MIMO system. To prove the main results, we analyze the edge behavior of the asymptotic ESD for $Y_t{Y}_t^{o}p$, and establish some sharp local laws on the resolvent of $Y_t{Y}_t^{o}p$. These results can be of independent interest, and used as useful inputs for many other problems regarding the spectral statistics of $Y_t$.

Recommendations

• Edge universality for deformed Wigner matrices
• Extremal eigenvalues and eigenvectors of deformed Wigner matrices
• On universality of local edge regime for the deformed Gaussian unitary ensemble
• Tracy-Widom at each edge of real covariance and MANOVA estimators
• High dimensional deformed rectangular matrices with applications in matrix denoising

Cites work

• A Dynamical Approach to Random Matrix Theory
• A necessary and sufficient condition for edge universality at the largest singular values of covariance matrices
• A necessary and sufficient condition for edge universality of Wigner matrices
• Almost sure localization of the eigenvalues in a Gaussian information plus noise model. Application to the spiked models.
• Analysis of the limiting spectral distribution of large dimensional information-plus-noise type matrices
• Averaging fluctuations in resolvents of random band matrices
• Bootstrap-based regularization for low-rank matrix estimation
• Edge statistics of large dimensional deformed rectangular matrices
• Edge universality for deformed Wigner matrices
• Edge universality of separable covariance matrices
• Estimating latent asset-pricing factors
• Estimation of a signal waveform from noisy data using low-rank approximation to a data matrix
• Finite sample approximation results for principal component analysis: A matrix perturbation approach
• High dimensional deformed rectangular matrices with applications in matrix denoising
• Improved Detection of Correlated Signals in Low-Rank-Plus-Noise Type Data Sets Using Informative Canonical Correlation Analysis (ICCA)
• Improved Subspace Estimation for Multivariate Observations of High Dimension: The Deterministic Signals Case
• Inference on low-rank data matrices with applications to microarray data
• Limiting Eigenvectors of Outliers for Spiked Information-Plus-Noise Type Matrices
• Local law and Tracy-Widom limit for sparse sample covariance matrices
• Local law for random Gram matrices
• Massive MIMO Systems With Non-Ideal Hardware: Energy Efficiency, Estimation, and Capacity Limits
• Minimax risk of matrix denoising by singular value thresholding
• On the empirical distribution of eigenvalues of large dimensional information-plus-noise-type matrices
• On the principal components of sample covariance matrices
• OptShrink: An Algorithm for Improved Low-Rank Signal Matrix Denoising by Optimal, Data-Driven Singular Value Shrinkage
• Optimal Shrinkage of Singular Values
• Random Matrix Theory and Wireless Communications
• Rate Optimal Denoising of Simultaneously Sparse and Low Rank Matrices
• Rigidity of eigenvalues of generalized Wigner matrices
• Robust high-dimensional factor models with applications to statistical machine learning
• Singular vector and singular subspace distribution for the matrix denoising model
• Singularities of the density of states of random Gram matrices
• Spectral statistics of Erdős-Rényi graphs II: eigenvalue spacing and the extreme eigenvalues
• Spectral statistics of Erdős-Rényi graphs. I: Local semicircle law
• Spectrum estimation for large dimensional covariance matrices using random matrix theory
• Spiked separable covariance matrices and principal components
• Testing hypotheses about the number of factors in large factor models
• The isotropic semicircle law and deformation of Wigner matrices
• The outliers of a deformed Wigner matrix
• The phase transition of matrix recovery from Gaussian measurements matches the minimax MSE of matrix denoising
• The singular values and vectors of low rank perturbations of large rectangular random matrices
• Tracy-Widom limit for the largest eigenvalue of a large class of complex sample covariance matrices
• Universality of covariance matrices

Cited in

(4)

• Tracy-Widom law for the extreme eigenvalues of large signal-plus-noise matrices
• Limiting distribution of the sample canonical correlation coefficients of high-dimensional random vectors
• Sample canonical correlation coefficients of high-dimensional random vectors with finite rank correlations
• Edge statistics of large dimensional deformed rectangular matrices

Describes a project that uses

Uses Software

• OptShrink

This page was built for publication: Edge statistics of large dimensional deformed rectangular matrices

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2079603)