Practical and powerful kernel-based change-point detection

TL;DR

This paper introduces a new kernel-based change-point detection framework that leverages high-dimensional data patterns, providing powerful, fast, and easy-to-use tests with superior performance demonstrated on real data.

Contribution

The paper proposes a novel kernel-based change-point detection method that improves power and control over false discoveries in high-dimensional data, with analytic significance approximations and practical R implementation.

Findings

Superior detection performance compared to existing methods

Analytic approximations enable fast significance testing

Effective application demonstrated on phone-call network data

Abstract

Change-point analysis plays a significant role in various fields to reveal discrepancies in distribution in a sequence of observations. While a number of algorithms have been proposed for high-dimensional data, kernel-based methods have not been well explored due to difficulties in controlling false discoveries and mediocre performance. In this paper, we propose a new kernel-based framework that makes use of an important pattern of data in high dimensions to boost power. Analytic approximations to the significance of the new statistics are derived and fast tests based on the asymptotic results are proposed, offering easy off-the-shelf tools for large datasets. The new tests show superior performance for a wide range of alternatives when compared with other state-of-the-art methods. We illustrate these new approaches through an analysis of a phone-call network data. All proposed methods are implemented in an R package KerSeg.