The development of an information criterion for Change-Point Analysis

TL;DR

This paper introduces a unified, information-based criterion for change-point analysis in time series, combining frequentist and information-theoretic methods to improve detection of state transitions.

Contribution

It develops a new, statistically principled change-point detection method that is parameter-free, prior-free, and applicable across various problems.

Findings

Reconciles frequentist and information-based approaches

Provides a parameter-free, prior-free testing method

Widely applicable to different change-point scenarios

Abstract

Change-point analysis is a flexible and computationally tractable tool for the analysis of times series data from systems that transition between discrete states and whose observables are corrupted by noise. The change-point algorithm is used to identify the time indices (change points) at which the system transitions between these discrete states. We present a unified information-based approach to testing for the existence of change points. This new approach reconciles two previously disparate approaches to Change-Point Analysis (frequentist and information-based) for testing transitions between states. The resulting method is statistically principled, parameter and prior free and widely applicable to a wide range of change-point problems.