A1443
Title: Change point detection in time series using mixed integer programming
Authors: Alexander Semenov - University of Florida (United States) [presenting]
Anton Skrobotov - Russian Presidential Academy of National Economy and Public Administration and SPBU (Russia)
Artem Prokhorov - University of Sydney (Australia)
Abstract: Recent advances in mixed-integer optimization (MIO) methods are used to develop a framework for the identification and estimation of structural breaks in time series. The framework requires a transformation of the classical structural break detection problem into a Mixed-Integer Quadratic Programming problem. MIO is capable of finding provably optimal solutions to this problem using a well-known optimization solver. The framework allows determining the unknown number of structural breaks. In addition to that, we demonstrate how to accommodate a specific required number of structural breaks, or a minimum required number of breaks. We demonstrate the effectiveness of our approach through extensive numerical experiments on synthetic and real-world data. We examine optimal and sub-optimal solutions to the problem, and the effect of tuning the parameters. We show how to choose the tuning parameters and compare our results with established econometric methods.
