Detecção de ponto de mudança nos parâmetros do modelo normal assimétrico: Algoritmo EM

Abstract

The change point problem emerged initially in quality control (PAGE, 1954) and before the introduction of the change point hypothesis associated with models, researchers faced difficulties in establishing a model for some datasets. In this work, we propose to study the change point problem in asymmetry context. We have developed a simple EM algorithm for maximum likelihood inference of the parameters of the skew-normal change point model. The Schwarz information criterion is proposed to locate a change point in the skew-normal model parameters. Moreover, the observed Fisher information matrix is analytically derived and some simulation studies and applications to real data sets are finally presented for the propose of illustrating the model and inferential results developed here.