A Model Fit Comparative Study of K-Component Mixture of One Parameter Univariate Distributions
Udochukwu Victor Echebiri *
Department of Statistics, Faculty of Physical Sciences, University of Benin, Benin, Nigeria.
Christogonus Ifeanyichukwu Ugoh
Department of Statistics, Faculty of Physical Sciences, Nnamdi Azikiwe University, Awka, Nigeria.
Emwinloghosa Kenneth Guobadia
Department of Administration, Federal Medical Centre, Asaba, Delta State, Nigeria.
Onaghise Andrew Isibor
Department of Statistics, Faculty of Physical Sciences, University of Benin, Benin, Nigeria.
Abayomi Omotayo
Department of Statistics, Faculty of Physical Sciences, University of Benin, Benin, Nigeria.
*Author to whom correspondence should be addressed.
Abstract
This is a comparative study on mixture distribution; where the study seeks to ascertain whether higher number of k-component mixtures could result to development of models that show better fits. In the performance comparison, special consideration was given to univariate one parameter distributions derived using mixture models, and the results show that distributions of higher k-mixture components relatively have greater propensity to exhibit better fit than the lesser mixture component distributions (k < 3).
Keywords: Mixture distribution, component mixtures, AIC, gamma distribution, model fit