ON ZERO INFLATED GENERALIZED POISSON-SUJATHA DISTRIBUTION AND ITS APPLICATIONS

Abstract

A new model called Zero Inflated generalized Poisson-Sujatha distribution (ZIGPSD) is proposed in this paper. Some characteristics of the model were derived and the maximum likelihood method was used to obtain the estimators of the parameters numerically. Through simulation and application to two datasets, the goodness-of-fit of the ZIGPSD was examined in comparison with the zero inflated Poisson (ZIP), Poisson and the new generalized Poisson-Sujatha (NGPS) models. The results showed that the ZIGPSD provides better fit compared to the other existing models considered in some cases of count data with excess zeros