Distributional stability of sparse inverse covariance matrix estimators

TL;DR

This paper examines the stability of sparse inverse covariance matrix estimators under data contamination, providing explicit bounds on distributional differences and analyzing implications for reliability in statistical inference.

Contribution

It introduces explicit local Lipschitz bounds for the distributional stability of sparse precision matrix estimators under contamination, a novel theoretical insight.

Findings

Derived explicit bounds for distributional differences under contamination

Analyzed stability of covariance matrix estimators and eigenvalues

Conducted numerical experiments confirming theoretical results

Abstract

Finding an approximation of the inverse of the covariance matrix, also known as precision matrix, of a random vector with empirical data is widely discussed in finance and engineering. In data-driven problems, empirical data may be ``contaminated''. This raises the question as to whether the approximate precision matrix is reliable from a statistical point of view. In this paper, we concentrate on a much-noticed sparse estimator of the precision matrix and investigate the issue from the perspective of distributional stability. Specifically, we derive an explicit local Lipschitz bound for the distance between the distributions of the sparse estimator under two different distributions (regarded as the true data distribution and the distribution of ``contaminated'' data). The distance is measured by the Kantorovich metric on the set of all probability measures on a matrix space. We also present analogous results for the standard estimators of the covariance matrix and its eigenvalues. Furthermore, we discuss several applications and conduct some numerical experiments.