House Price Changes: A Perspective of Inhomogeneous Multiple Structural Breaks

Abstract

This article proposes an inhomogeneous multiple structural breaks model, with unknown mean changes or variance changes in observations. The asymptotic properties of the likelihood ratio statistics are demonstrated, which converge to a Gumbel distribution. Furthermore, the rule of information criterion and Bayesian model selection are used to detect inhomogeneous multiple change points through the binary segmentation and random interval technique. Extensive simulation experiments are provided to illustrate the promising performance of our method. The real residential property prices for Australia and China are employed in this model for empirical analysis, which illustrates the performance of the proposed model.