A Poisson Regression Model For Analysis of Censored Count Data with Excess Zeroes

Abstract

Typically, a Poisson regression model is assumed for count data. In many cases, there are many zeros in the dependent variable, therefore the mean is not equal to the variance value of the dependent variable. Thus, we suggest using a hurdle and zero-inflated Poisson regression model. Furthermore, the response variable in such cases is censored for some values. In this paper, a censored hurdle Poisson regression model and a censored zero-inflated Poisson regression model will be discussed to handle the overdispersion problem when there are excess zeros in the response variable. The estimation of regression parameters using the maximum likelihood method is discussed and the goodness-of-fit statistics for the regression model are examined. An example and a simulation will be used to compare the censored hurdle Poisson regression model with the censored zero-inflated Poisson regression model in terms of the parameter estimation, standard errors and the goodness-of-fit statistics.