Extreme eigenvalues of large-dimensional spiked Fisher matrices with application

DOI10.1214/16-AOS1463MaRDI QIDQ524466zbMATH OpenOpenAlexFDO

Authors Qinwen Wang, Jian-Feng Yao

Publication date 2 May 2017

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1504.05087, https://projecteuclid.org/euclid.aos/1487667628

zbMATH Keywords

central limit theoremextreme eigenvaluehigh-dimensional data analysislarge-dimensional Fisher matricesphase transitionsignal detectionspiked Fisher matrixspiked population model

Mathematics Subject Classification ID

Central limit and other weak theorems (60F05) Estimation in multivariate analysis (62H12)

Abstract: Consider two $p$-variate populations, not necessarily Gaussian, with covariance matrices $Sigm{a}_1$ and $Sigm{a}_2$, respectively, and let $S_1$ and $S_2$ be the sample covariances matrices from samples of the populations with degrees of freedom $T$ and $n$, respectively. When the difference $Delta$ between $Sigm{a}_1$ and $Sigm{a}_2$ is of small rank compared to $p,T$ and $n$, the Fisher matrix $F=S_2^{-1}{S}_1$ is called a {em spiked Fisher matrix}. When $p,T$ and $n$ grow to infinity proportionally, we establish a phase transition for the extreme eigenvalues of $F$: when the eigenvalues of $Delta$ ({em spikes}) are above (or under) a critical value, the associated extreme eigenvalues of the Fisher matrix will converge to some point outside the support of the global limit (LSD) of other eigenvalues; otherwise, they will converge to the edge points of the LSD. Furthermore, we derive central limit theorems for these extreme eigenvalues of the spiked Fisher matrix. The limiting distributions are found to be Gaussian if and only if the corresponding population spike eigenvalues in $Delta$ are {em simple}. Numerical examples are provided to demonstrate the finite sample performance of the results. In addition to classical applications of a Fisher matrix in high-dimensional data analysis, we propose a new method for the detection of signals allowing an arbitrary covariance structure of the noise. Simulation experiments are conducted to illustrate the performance of this detector.

Recommendations

• Limiting laws for extreme eigenvalues of large-dimensional spiked Fisher matrices with a divergent number of spikes
• The limits of the sample spiked eigenvalues for a high-dimensional generalized Fisher matrix and its applications
• CLT for eigenvalue statistics of large-dimensional general Fisher matrices with applications
• Statistical inference for principal components of spiked covariance matrices
• Asymptotic joint distribution of extreme eigenvalues and trace of large sample covariance matrix in a generalized spiked population model

Cited in

(19)

• Canonical correlation coefficients of high-dimensional Gaussian vectors: finite rank case
• Random matrix-improved estimation of covariance matrix distances
• Limiting laws for extreme eigenvalues of large-dimensional spiked Fisher matrices with a divergent number of spikes
• Random matrix theory and its applications
• Limiting distribution of the sample canonical correlation coefficients of high-dimensional random vectors
• CLT for spiked eigenvalues of a sample covariance matrix from high-dimensional Gaussian mean mixtures
• Testing in high-dimensional spiked models
• The limits of the sample spiked eigenvalues for a high-dimensional generalized Fisher matrix and its applications
• Order Determination for Spiked Type Models
• CLT for eigenvalue statistics of large-dimensional general Fisher matrices with applications
• Detection of weak signals in high-dimensional complex-valued data
• Approximation of the power functions of Roy's largest root test under general spiked alternatives
• Sample canonical correlation coefficients of high-dimensional random vectors with finite rank correlations
• Order determination for spiked-type models with a divergent number of spikes
• Spiked multiplicative random matrices and principal components
• A CLT for the LSS of large-dimensional sample covariance matrices with diverging spikes
• Spiked eigenvalues of noncentral Fisher matrix with applications
• Invariance principle and CLT for the spiked eigenvalues of large-dimensional Fisher matrices and applications
• Inference on testing the number of spikes in a high-dimensional generalized spiked Fisher matrix

This page was built for publication: Extreme eigenvalues of large-dimensional spiked Fisher matrices with application

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q524466)