Comparison of Mixture and Classification Maximum Likelihood Approaches in Poiss...

• Author(s): Faria, Susana ; Soromenho, Gilda
• Date: 2008
• Persistent ID: http://hdl.handle.net/10451/4713
• Origin: Repositório da Universidade de Lisboa
• Subject(s): Simulation study; EM algorithm; Mixture Poisson Regression Models; Classification EM algorithm
• Document Type: Conference Object
• Language: English

Description

In this work, we propose to compare two algorithms to compute maximum likelihood estimators of the parameters of a mixture Poisson regression models. To estimate these parameters, we may use the EM algorithm in a mixture approach or the CEM algorithm in a classification approach. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets in a target number of iterations. Simulation results show that the CEM algorithm is a good alternative to the EM algorithm for fitting Poisson mixture regression models, having the advantage of converging more quickly.