Saved in:
Bibliographic Details
Main Authors: Szwagier, Tom, Olikier, Guillaume, Pennec, Xavier
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.10110
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913938250137600
author Szwagier, Tom
Olikier, Guillaume
Pennec, Xavier
author_facet Szwagier, Tom
Olikier, Guillaume
Pennec, Xavier
contents Covariance estimation is a central problem in statistics. An important issue is that there are rarely enough samples $n$ to accurately estimate the $p (p+1) / 2$ coefficients in dimension $p$. Parsimonious covariance models are therefore preferred, but the discrete nature of model selection makes inference computationally challenging. In this paper, we propose a relaxation of covariance parsimony termed "eigengap sparsity" and motivated by the good accuracy-parsimony tradeoffs of eigenvalue-equalization in covariance matrices. This penalty can be included in a penalized-likelihood framework that we propose to solve with a projected gradient descent on a monotone cone. The algorithm turns out to resemble an isotonic regression of mutually-attracted sample eigenvalues, drawing an interesting link between covariance parsimony and shrinkage.
format Preprint
id arxiv_https___arxiv_org_abs_2504_10110
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Eigengap Sparsity for Covariance Parsimony
Szwagier, Tom
Olikier, Guillaume
Pennec, Xavier
Methodology
Covariance estimation is a central problem in statistics. An important issue is that there are rarely enough samples $n$ to accurately estimate the $p (p+1) / 2$ coefficients in dimension $p$. Parsimonious covariance models are therefore preferred, but the discrete nature of model selection makes inference computationally challenging. In this paper, we propose a relaxation of covariance parsimony termed "eigengap sparsity" and motivated by the good accuracy-parsimony tradeoffs of eigenvalue-equalization in covariance matrices. This penalty can be included in a penalized-likelihood framework that we propose to solve with a projected gradient descent on a monotone cone. The algorithm turns out to resemble an isotonic regression of mutually-attracted sample eigenvalues, drawing an interesting link between covariance parsimony and shrinkage.
title Eigengap Sparsity for Covariance Parsimony
topic Methodology
url https://arxiv.org/abs/2504.10110