Tracy-Widom Distribution for the Largest Eigenvalue of Real Sample Covariance Matrices with General Population

TL;DR

This paper establishes that the largest eigenvalue of certain real sample covariance matrices converges to the Tracy-Widom distribution under broad conditions, extending previous results to more general population matrices.

Contribution

It proves the Tracy-Widom law for the largest eigenvalue of real sample covariance matrices with general population matrices in the sub-critical regime, under broad distributional assumptions.

Findings

Largest eigenvalue follows Tracy-Widom distribution asymptotically

Results hold for non-Gaussian entries with subexponential decay

Applicable to general positive-definite population matrices

Abstract

We consider sample covariance matrices of the form $\mathcal{Q}=(\Sigma^{1/2}X)(\Sigma^{1/2} X)^*$, where the sample $X$ is an $M\times N$ random matrix whose entries are real independent random variables with variance $1/N$ and where $\Sigma$ is an $M\times M$ positive-definite deterministic matrix. We analyze the asymptotic fluctuations of the largest rescaled eigenvalue of $\mathcal{Q}$ when both $M$ and $N$ tend to infinity with $N/M\to d\in(0,\infty)$. For a large class of populations $\Sigma$ in the sub-critical regime, we show that the distribution of the largest rescaled eigenvalue of $\mathcal{Q}$ is given by the type-1 Tracy-Widom distribution under the additional assumptions that (1) either the entries of $X$ are i.i.d. Gaussians or (2) that $\Sigma$ is diagonal and that the entries of $X$ have a subexponential decay.