Models for Overdispersion Count Data with Generalized Distribution: An Application to Parasites Intensity

The Poisson regression model is widely used for count data. This assumes equidispersion. In practice, equidispersion seldom reflected in However, real-life data, the variance usually exceeds mean. situation known as overdispersion. Negative binomial distribution and other mix models are often to overdispersion Another extension of negative another data univariate generalized Waring. addition, developed by Famoye can be analysis When contains a large number zeros, it necessary use zero-inflated models. this study, different emphasized excessive zeros For purpose, real set was analysed with model, Famoye, Waring foregoing Log-likelihood, Akaike information criterion, Bayes Vuong statistics were comparisons.