On an Alternative to Discrete Pareto Model

Abstract

Here we introduce a novel and versatile discrete distribution as an alternative to discrete Pareto distribution and derive some noteworthy distributional properties. The proposed distribution has some interesting characteristics, such as decreasing hazard rate, unimodality, overdispersion and infinite divisibility. Some characterizations are also obtained. We also consider different estimation procedures for estimating the model parameters, namely: the maximum likelihood, least squares, weighted least squares, Cramer-von Mises methods and compare them using a comprehensive simulation study. Lastly, to illustrate the practical utility of the proposed model, we analyze four real datasets.