Central limit theorems for functionals of large sample covariance matrix and mean vector in matrix-variate location mixture of normal distributions

TL;DR

This paper establishes central limit theorems for functionals involving the sample covariance matrix and mean vector in high-dimensional settings with matrix-variate normal mixture distributions, covering cases with non-invertible covariance matrices.

Contribution

It derives new asymptotic distributions for products of sample covariance matrices and mean vectors under large-dimensional regimes with matrix-variate normal mixtures.

Findings

CLTs for the product of sample covariance matrix and mean vector.

CLTs for the product of inverse sample covariance matrix and mean vector.

Results applicable when the covariance matrix may be non-invertible.

Abstract

In this paper we consider the asymptotic distributions of functionals of the sample covariance matrix and the sample mean vector obtained under the assumption that the matrix of observations has a matrix-variate location mixture of normal distributions. The central limit theorem is derived for the product of the sample covariance matrix and the sample mean vector. Moreover, we consider the product of the inverse sample covariance matrix and the mean vector for which the central limit theorem is established as well. All results are obtained under the large-dimensional asymptotic regime where the dimension $p$ and the sample size $n$ approach to infinity such that $p/n\to c\in[0 , +\infty)$ when the sample covariance matrix does not need to be invertible and $p/n\to c\in [0, 1)$ otherwise.