Comparing discrete Pareto populations under a fixed effects model

Abstract

The discrete Pareto distribution can be considered as a lifetime distribution and then is widely used in practice. It follows the power law tails property which makes it as a candidate model for natural phenomena. This paper deals with comparison of discrete Pareto populations by proposing a non-linear fixed effects model. Estimators for the factor effects are derived in explicit expressions. Stochastic properties of the estimators are studied in details. A test for assessing the homogeneity of populations is proposed. Illustrative examples are also given. The proposed model is an alternative model for analyzing data sets in which the linear models have poor performance.