Poisson Regression
HI all,
Let's say that I am modeling number of claim count an insurance company have. I vaguely remember that if the deviacne/DF is approx 1.000 (in SAS Genmod) and we can safely assume the data fits Poisson well. In the case where the data doesn't fit Poisson, we can throw in the command "scale=deviance or pearson" to "fix" it. What is this command is doing? any explanation is appreciated. Ideally I want to be able to calculate by hand to fully understand it. Thanks in advance.
