Spiked eigenvalues of noncentral Fisher matrix with applications

04/10/2021

∙

by   Xiaozhuo Zhang, et al.

∙

0

∙

share

In this paper, we investigate the asymptotic behavior of spiked eigenvalues of the noncentral Fisher matrix defined by 𝐅_p=𝐂_n(𝐒_N)^-1, where 𝐂_n is a noncentral sample covariance matrix defined by (Ξ+𝐗)(Ξ+𝐗)^*/n and 𝐒_N=𝐘𝐘^*/N. The matrices 𝐗 and 𝐘 are two independent Gaussian arrays, with respective p× n and p× N and the Gaussian entries of them are independent and identically distributed (i.i.d.) with mean 0 and variance 1. When p, n, and N grow to infinity proportionally, we establish a phase transition of the spiked eigenvalues of 𝐅_p. Furthermore, we derive the central limiting theorem (CLT) for the spiked eigenvalues of 𝐅_p. As an accessory to the proof of the above results, the fluctuations of the spiked eigenvalues of 𝐂_n are studied, which should have its own interests. Besides, we develop the limits and CLT for the sample canonical correlation coefficients by the results of the spiked noncentral Fisher matrix and give three consistent estimators, including the population spiked eigenvalues and the population canonical correlation coefficients.