On the Finite Mixture Modelling via Normal Mean-variance Birnbaum-Saunders Distribution

ýThis paper presents a new finite mixture model using the normal mean-varianceý ýmixture of Birnbaum-Saunders distributioný. ýThe proposed model is multimodal with widerý ýranges of skewness and kurtosisý. ýMoreoverý, ýit is useful for modeling highly asymmetric data in various theoretical and applied statistical problemsý. ýThe maximum likelihoodý ýestimates of the parameters of the model are computed iteratively by feasibleý ýEM algorithmý. ýTo illustrate the finite sample properties and performance of theý ýestimatorsý, ýwe conduct a simulation study and illustrate the usefulness of the new model by analyzing a real datasetý.