The new flexible generalized class of distributions with applications to different fields

Abstract

We propose a new generator which has been used as a generalized class, and can also be helpful in generating new flexible generalized classes of distributions for continuous random variable. The newly proposed generator-cum-generalized class does not involve any extra parameter, and its functional form plays an important role along with baseline models to develop flexible models. In literature, such classes had been reported as Marshall-Olkin G-class, exponentiated G-class, Transmuted G-class, exponentiated generalized G-class, and Flexible G-class. So, any parent (or baseline) model can be substituted in the proposed class which has no extra burden on the parameters of the class. Some needful characteristics of the newly proposed class are obtained. Furthermore, a special model of the class, that is, the flexible Kumaraswamy distribution is considered and its properties are reported. The parameters estimation is dealt through the method of maximum likelihood, and a simulation is carried out to assess the performance of model’s parameters. Four real-life data sets are analyzed to show usefulness of the proposed model in comparison to some well-established competitive models. It is found that the proposed model yields low values of the goodness-of-fit statistics as compared to the other models, and hence our proposed model performed better as compared to others on these four data sets.