The methods for estimating the number of components in mixture distribution models

Mixture distribution models can be used for modelling of non-homogeneousor heteregeneous structure of a population with respect to the interestedcharacteristics.In this study, the methods for estimating the number of components inmixture distribution models are examined. For this purpose in the first chapter, someinformations of mixture distribution models are given and in the second chapter abrief history of mixture distribution models are given. In the third chapter, generalinformations about formulations of mixture distribution models are given. In thefourth chapter, the graphical methods for estimating the number of components inmixture distribution models are analysed. In the fifth chapter, applying likelihoodratio test statistic to estimate the number of compnents in mixture distribution modelsare given. In the sixth chapter, information criterion in model selection for mixturedistribution models are examined. In seventh chapter, estimating parametres inmixture distribution models are given. In eighth chapter, explained methods areapplied on a sample data set. Finally, results and discussions are givenKey Words: Determining the number of components, Expectation andmaximization algorithm, Information criterion, Mixture distribution model,Paramater Estimation.