Lagged Exact Bayesian Online Changepoint Detection with Parameter Estimation

TL;DR

This paper introduces LEXO, an improved Bayesian online changepoint detection algorithm that enhances stability and accuracy by incorporating lagged inference and parameter estimation, outperforming previous methods in simulations and real data.

Contribution

The paper proposes LEXO, a novel lagged Bayesian changepoint detection algorithm that improves stability and parameter estimation over existing methods like EXO.

Findings

LEXO provides more stable changepoint detection.

Parameter estimates from LEXO have lower MSE.

LEXO performs well on real-world data examples.

Abstract

Identifying changes in the generative process of sequential data, known as changepoint detection, has become an increasingly important topic for a wide variety of fields. A recently developed approach, which we call EXact Online Bayesian Changepoint Detection (EXO), has shown reasonable results with efficient computation for real time updates. The method is based on a \textit{forward} recursive message-passing algorithm. However, the detected changepoints from these methods are unstable. We propose a new algorithm called Lagged EXact Online Bayesian Changepoint Detection (LEXO) that improves the accuracy and stability of the detection by incorporating $\ell$-time lags to the inference. The new algorithm adds a recursive \textit{backward} step to the forward EXO and has computational complexity linear in the number of added lags. Estimation of parameters associated with regimes is also developed. Simulation studies with three common changepoint models show that the detected changepoints from LEXO are much more stable and parameter estimates from LEXO have considerably lower MSE than EXO. We illustrate applicability of the methods with two real world data examples comparing the EXO and LEXO.