Sparse Multivariate Linear Regression with Strongly Associated Response Variables

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ham, Daeyoung, Price, Bradley S., Rothman, Adam J.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911101760831488
author Ham, Daeyoung
Price, Bradley S.
Rothman, Adam J.
author_facet Ham, Daeyoung
Price, Bradley S.
Rothman, Adam J.
contents We propose new methods for multivariate linear regression when the regression coefficient matrix is sparse and the error covariance matrix is dense. We assume that the error covariance matrix has equicorrelation across the response variables. Two procedures are proposed: one is based on constant marginal response variance (compound symmetry), and the other is based on general varying marginal response variance. Two approximate procedures are also developed for high dimensions. We propose an approximation to the Gaussian validation likelihood for tuning parameter selection. Extensive numerical experiments illustrate when our procedures outperform relevant competitors as well as their robustness to a moderate degree of model misspecification.
format Preprint
id arxiv_https___arxiv_org_abs_2410_10025
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Sparse Multivariate Linear Regression with Strongly Associated Response Variables
Ham, Daeyoung
Price, Bradley S.
Rothman, Adam J.
Methodology
We propose new methods for multivariate linear regression when the regression coefficient matrix is sparse and the error covariance matrix is dense. We assume that the error covariance matrix has equicorrelation across the response variables. Two procedures are proposed: one is based on constant marginal response variance (compound symmetry), and the other is based on general varying marginal response variance. Two approximate procedures are also developed for high dimensions. We propose an approximation to the Gaussian validation likelihood for tuning parameter selection. Extensive numerical experiments illustrate when our procedures outperform relevant competitors as well as their robustness to a moderate degree of model misspecification.
title Sparse Multivariate Linear Regression with Strongly Associated Response Variables
topic Methodology
url https://arxiv.org/abs/2410.10025