Change Point Detection in the Mean of High-Dimensional Time Series Data under Dependence

TL;DR

This paper introduces a new method for detecting change points in the mean of high-dimensional time series data that accounts for complex dependence structures, with proven asymptotic properties and demonstrated effectiveness on real fMRI data.

Contribution

A novel change point detection procedure that handles dependence in high-dimensional time series and accurately estimates boundary change points.

Findings

Robust detection performance in simulations.

Effective application to fMRI data.

Theoretical guarantees under mild conditions.

Abstract

High-dimensional time series are characterized by a large number of measurements and complex dependence, and often involve abrupt change points. We propose a new procedure to detect change points in the mean of high-dimensional time series data. The proposed procedure incorporates spatial and temporal dependence of data and is able to test and estimate the change point occurred on the boundary of time series. We study its asymptotic properties under mild conditions. Simulation studies demonstrate its robust performance through the comparison with other existing methods. Our procedure is applied to an fMRI dataset.