Testing for structural change in a mixture of linear regressions

Abstract

Over the last years, several structural change tests have been extended to monitoring of linear regression models where new data arrive over time. In this work, we derive a new procedure for the problem testing for structural change in a mixture of linear regressions. In this procedure, the parameters of a mixture of linear regression model are estimated from all available data (initial data plus newly arrived data) and compared to the estimates based only on initial data. The procedure is illustrated through a simulation study on both simulated and real data sets.