Inference on Structural Breaks using Information Criteria

This paper investigates the usefulness of information criteria for inference on the number of structural breaks in a standard linear regression model. In particular, we propose a modified penalty function for such criteria based on theoretical arguments, which implies each break is equivalent to estimation of three individual regression coefficients. A Monte Carlo analysis compares information criteria to sequential testing, with the modified BIC and HQIC criteria performing well overall, for DGPs both without and with breaks. The methods are also used to examine changes in Euro area monetary policy between 1971 and 2007.