Comparison of EM and SEM Algorithms in Poisson Regression Models: A Simulation Study

• Published in: Communications in statistics. Simulation and computation Vol. 41; no. 4; pp. 497 - 509
• Main Authors: Faria, Susana, Soromenho, Gilda
• Format: Journal Article
• Language: English
• Published: Colchester Taylor & Francis Group 01.04.2012; Taylor & Francis; Taylor & Francis Ltd
• Subjects: Algorithms; Computation; Computer simulation; Economic models; EM algorithm; Estimates; Exact sciences and technology; Linear inference, regression; Mathematical models; Mathematics; Maximum likelihood estimates; Maximum likelihood estimation; Mixture Poisson regression models; Numerical analysis; Numerical analysis. Scientific computation; Numerical methods in probability and statistics; Parametric inference; Probability and statistics; Regression; Sciences and techniques of general use; Simulation; Simulation study; Statistics; Stochastic EM algorithm; Stochasticity; Mixed distribution; 62G05; Stochastic model; Maximum likelihood estimation; Mixture Poisson regression models; Stochastic EM algorithm; Statistical simulation; Statistical method; Statistical regression; Hypothesis test; Statistical test; Simulation study; Regression model; Numerical simulation; Maximum likelihood; Simulation model; 62J 12; EM algorithm; Physical Sciences
• ISSN: 0361-0918; 1532-4141; 1532-4141
• DOI: 10.1080/03610918.2011.594534

In this work, we propose to compare two algorithms to compute maximum likelihood estimates for the parameters of a mixture Poisson regression model: the EM algorithm and the Stochastic EM algorithm. The comparison of the two procedures was done through a simulation study of the performance of these approaches on simulated data sets and real data sets. Simulation results show that the choice of the approach depends essentially on the overlap of the regression lines. In the real data case, we show that the Stochastic EM algorithm resulted in model estimates that best fit the regression model.