A0564
Title: A family of distributions for modelling dispersion in count data
Authors: Shin Zhu Sim - Universiti Tunku Abdul Rahman (Malaysia)
Seng Huat Ong - University of Malaya (Malaysia) [presenting]
Abstract: A family of distributions is considered which has the flexibility to cater for under- , equi- and over- dispersion in count data. The motivation to introduce this family is because it has a simple finite mixture structure and which facilitates its application and analysis. Some basic and probabilistic properties have been derived. Tests of hypothesis for equi-dispersion and simulation study of their power, parameter estimation by maximum likelihood and a probability generating function based methods have been considered. The utility of the distributions are illustrated by application to real data sets which exhibit under, equi- and over dispersion. It is shown that the distribution fits better than the well-known generalized Poisson and COM-Poisson distributions.
