Saved in:
Bibliographic Details
Main Authors: Yang, Xuzhi, Wang, Tengyao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.09098
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913234421809152
author Yang, Xuzhi
Wang, Tengyao
author_facet Yang, Xuzhi
Wang, Tengyao
contents Composite quantile regression has been used to obtain robust estimators of regression coefficients in linear models with good statistical efficiency. By revealing an intrinsic link between the composite quantile regression loss function and the Wasserstein distance from the residuals to the set of quantiles, we establish a generalization of the composite quantile regression to the multiple-output settings. Theoretical convergence rates of the proposed estimator are derived both under the setting where the additive error possesses only a finite $\ell$-th moment (for $\ell > 2$) and where it exhibits a sub-Weibull tail. In doing so, we develop novel techniques for analyzing the M-estimation problem that involves Wasserstein-distance in the loss. Numerical studies confirm the practical effectiveness of our proposed procedure.
format Preprint
id arxiv_https___arxiv_org_abs_2402_09098
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multiple-output composite quantile regression through an optimal transport lens
Yang, Xuzhi
Wang, Tengyao
Statistics Theory
Composite quantile regression has been used to obtain robust estimators of regression coefficients in linear models with good statistical efficiency. By revealing an intrinsic link between the composite quantile regression loss function and the Wasserstein distance from the residuals to the set of quantiles, we establish a generalization of the composite quantile regression to the multiple-output settings. Theoretical convergence rates of the proposed estimator are derived both under the setting where the additive error possesses only a finite $\ell$-th moment (for $\ell > 2$) and where it exhibits a sub-Weibull tail. In doing so, we develop novel techniques for analyzing the M-estimation problem that involves Wasserstein-distance in the loss. Numerical studies confirm the practical effectiveness of our proposed procedure.
title Multiple-output composite quantile regression through an optimal transport lens
topic Statistics Theory
url https://arxiv.org/abs/2402.09098