Multiple Change Point Detection and Validation in Autoregressive Time Series Data

TL;DR

This paper introduces a method for detecting and validating multiple change points in autoregressive time series data, combining likelihood ratio scans with spectral discrimination tests for improved accuracy.

Contribution

It proposes a novel two-step approach that first identifies potential change points and then validates them using spectral tests, enhancing change point detection in autoregressive models.

Findings

Effective detection of change points demonstrated through numerical studies

Method outperforms some existing techniques in accuracy

Applicable to various scenarios with different data structures

Abstract

It is quite common that the structure of a time series changes abruptly. Identifying these change points and describing the model structure in the segments between these change points is of interest. In this paper, time series data is modelled assuming each segment is an autoregressive time series with possibly different autoregressive parameters. This is achieved using two main steps. The first step is to use a likelihood ratio scan based estimation technique to identify these potential change points to segment the time series. Once these potential change points are identified, modified parametric spectral discrimination tests are used to validate the proposed segments. A numerical study is conducted to demonstrate the performance of the proposed method across various scenarios and compared against other contemporary techniques.