UM TUTORIAL SOBRE ESTIMAÇÃO EM MODELOS DE MISTURA

Main Article Content

Laisa Ribeiro ZUCARELI
Erlandson Ferreira SARAIVA
Adriano Kamimura SUZUKI

Abstract

In the last years, there has been a growing interest in modeling using mixture models due its exibility. One of the main issues in the estimation procedure of its parameters is the determination of the number of components. In this paper, through a Monte Carlo simulation study, we compared the performance of the model selection criteria AIC, AICc and BIC for determining the number of components of a mixture model. The results obtained show a complementarity between the criteria and that these should be used with some care, since they present percentages of success less than 70% in most of the studied cases.

Article Details

How to Cite
ZUCARELI, L. R., SARAIVA, E. F., & SUZUKI, A. K. (2018). UM TUTORIAL SOBRE ESTIMAÇÃO EM MODELOS DE MISTURA. Brazilian Journal of Biometrics, 36(4), 968–997. https://doi.org/10.28951/rbb.v36i4.331
Section
Articles

Most read articles by the same author(s)