A SYSTEMATIC PROCEDURE FOR MODELING ASYMMETRICAL DATA

In this study, we develop a framework for selecting a model that is appropriate for the asymmetric data. We cannot predict with certainty that the data under analysis will always have a symmetric probability distribution. In many practical situations, the data under study may follow skewed probability distributions. It is challenging to model these kinds of skewed data sets. An effective way of modeling these types of data sets is using a finite mixture of probability distributions. In order to identify a better mixture model, we also fitted the composite model and the T-X family to compare the efficacy of the finite mixture models in terms of goodness of fit tests and information criteria. Further, the parameter estimation for the proposed model as well as other models selected for comparison the maximum likelihood estimation method is used. The statistical characteristics of the suggested mixture model are also analyzed in detail. Received: May 16, 2024Revised: July 2, 2024Accepted: July 19, 2024