Performance Analysis of Tyler's Covariance Estimator
Statistics Theory
2015-06-18 v4 Statistics Theory
Abstract
This paper analyzes the performance of Tyler's M-estimator of the scatter matrix in elliptical populations. We focus on the non-asymptotic setting and derive the estimation error bounds depending on the number of samples n and the dimension p. We show that under quite mild conditions the squared Frobenius norm of the error of the inverse estimator decays like p^2/n with high probability.
Keywords
Cite
@article{arxiv.1401.6926,
title = {Performance Analysis of Tyler's Covariance Estimator},
author = {Ilya Soloveychik and Ami Wiesel},
journal= {arXiv preprint arXiv:1401.6926},
year = {2015}
}