<![CDATA[Regression analysis is one method to determine and test the causality relationship (cause-effect) between the dependent variable (Y) with the independent variables (X). In general, regression analysis is used to analyze non-free variable data in the form of continuous data and normal distribution. However, in some applications, non-free variable data to be analyzed in the form of discrete data and not normally distributed. One of the regression models that can be used to analyze the relationship between the dependent variable (Y) in the form of discrete data is Poisson regression model whose dependent variable is Poisson distributed. Poisson regression has the assumption of equidispersion that is the condition in which the mean and variance values of the dependent variable are equal, but sometimes there is an assumption violation, where the value of variance is greater than the so-called overdispersion value, so to overcome it can be used one of the extensions of the regression model Poisson is Poisson regression model generalized, this is because the assumption does not require the same mean value with the value of variance. The purpose of this study is how to estimate the Poisson regression model and Poisson regression model generalized I and explain how the generalized Poisson regression model I in overcoming the overdispersion in Poisson regression.]]>
Copyrights © 2015
