Bayesian Indicator-Saturated Regression for Climate Policy Evaluation

TL;DR

This paper introduces a Bayesian indicator-saturated regression framework for detecting structural breaks in longitudinal data, improving climate policy evaluation accuracy, especially when multiple breaks are present.

Contribution

It presents a unified probabilistic approach with a spike-and-slab prior for structural break detection, outperforming frequentist methods in high-break environments.

Findings

Outperforms frequentist approaches in simulations

Effectively detects multiple structural breaks

Successfully applied to climate policy data in Europe

Abstract

Structural break identification methods are an important tool for evaluating the effectiveness of climate change mitigation policies. In this paper, we introduce a unified probabilistic framework for detecting structural breaks with unknown timing and arbitrary sequence in longitudinal data. The proposed Bayesian setup uses indicator-saturated regression and a spike-and-slab prior with an inverse-moment density as the slab component to ensure model selection consistency. Simulation results show that the method outperforms comparable frequentist approaches, particularly in environments with a high probability of structural breaks. We apply the framework to identify and evaluate the effects of climate policies in the European road transport sector.