Approximate simulation-free Bayesian inference for multiple changepoint models with dependence within segments

TL;DR

This paper introduces simulation-free Bayesian methods for multiple changepoint detection in dependent data, using hierarchical Gaussian Markov random fields and efficient approximations to reduce computational costs.

Contribution

It develops novel, simulation-free Bayesian inference techniques for changepoint models with within-segment dependence, improving computational efficiency and modeling flexibility.

Findings

Methods effectively detect changepoints in dependent data

Models outperform independent-data approaches in real data analysis

Significant computational savings achieved with proposed approximations

Abstract

This paper proposes approaches for the analysis of multiple changepoint models when dependency in the data is modelled through a hierarchical Gaussian Markov random field. Integrated nested Laplace approximations are used to approximate data quantities, and an approximate filtering recursions approach is proposed for savings in compuational cost when detecting changepoints. All of these methods are simulation free. Analysis of real data demonstrates the usefulness of the approach in general. The new models which allow for data dependence are compared with conventional models where data within segments is assumed independent.