Abstract

Poisson Regression, Negatif Binomial Regression, Zero-Inflated Poisson Regression and Zero-Inflated Negative Binomial Regression methods are user to modelling count data with excess zeros and / or over dispersion. In this study, it is considered that the number of complaints received from customers in any service sector is influenced by gender, age, education and experience variables. These data are the analyzed to assess zero-inflated procedures. Also, Akaike Information Criterion is used to evaluate the regression models. In practice, it is determined which model is suitable for each month of the year 2016 and interpretations are made about parameter estimations of appropriate models.