Simultaneous Detection and Localization of Mean and Covariance Changes in High Dimensions

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cui, Junfeng, Pan, Guangming, Wang, Guanghui, Zou, Changliang
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914008741707776
author Cui, Junfeng
Pan, Guangming
Wang, Guanghui
Zou, Changliang
author_facet Cui, Junfeng
Pan, Guangming
Wang, Guanghui
Zou, Changliang
contents Existing methods for high-dimensional changepoint detection and localization typically focus on changes in either the mean vector or the covariance matrix separately. This separation reduces detection power and localization accuracy when both parameters change simultaneously. We propose a simple yet powerful method that jointly monitors shifts in both the mean and covariance structures. Under mild conditions, the test statistics for detecting these shifts jointly converge in distribution to a bivariate standard normal distribution, revealing their asymptotic independence. This independence enables the combination of the individual p-values using Fisher's method, and the development of an adaptive p-value-based estimator for the changepoint. Theoretical analysis and extensive simulations demonstrate the superior performance of our method in terms of both detection power and localization accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19523
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Simultaneous Detection and Localization of Mean and Covariance Changes in High Dimensions
Cui, Junfeng
Pan, Guangming
Wang, Guanghui
Zou, Changliang
Statistics Theory
Methodology
62G10 (Primary) 62G20, 62H15 (Secondary)
Existing methods for high-dimensional changepoint detection and localization typically focus on changes in either the mean vector or the covariance matrix separately. This separation reduces detection power and localization accuracy when both parameters change simultaneously. We propose a simple yet powerful method that jointly monitors shifts in both the mean and covariance structures. Under mild conditions, the test statistics for detecting these shifts jointly converge in distribution to a bivariate standard normal distribution, revealing their asymptotic independence. This independence enables the combination of the individual p-values using Fisher's method, and the development of an adaptive p-value-based estimator for the changepoint. Theoretical analysis and extensive simulations demonstrate the superior performance of our method in terms of both detection power and localization accuracy.
title Simultaneous Detection and Localization of Mean and Covariance Changes in High Dimensions
topic Statistics Theory
Methodology
62G10 (Primary) 62G20, 62H15 (Secondary)
url https://arxiv.org/abs/2508.19523