174
v1v2 (latest)

Tracy-Widom distribution for the edge eigenvalues of elliptical model

Information and Inference A Journal of the IMA (JIII), 2023
Abstract

In this paper, we study the largest eigenvalues of sample covariance matrices with elliptically distributed data. We consider the sample covariance matrix Q=YY,Q=YY^*, where the data matrix YRp×nY \in \mathbb{R}^{p \times n} contains i.i.d. pp-dimensional observations yi=ξiTui,  i=1,,n.\mathbf{y}_i=\xi_iT\mathbf{u}_i,\;i=1,\dots,n. Here ui\mathbf{u}_i is distributed on the unit sphere, ξiξ\xi_i \sim \xi is independent of ui\mathbf{u}_i and TT=ΣT^*T=\Sigma is some deterministic matrix. Under some mild regularity assumptions of Σ,\Sigma, assuming ξ2\xi^2 has bounded support and certain proper behavior near its edge so that the limiting spectral distribution (LSD) of QQ has a square decay behavior near the spectral edge, we prove that the Tracy-Widom law holds for the largest eigenvalues of QQ when pp and nn are comparably large.

View on arXiv
Comments on this paper