[R] Obtain Log-Likelihood for ml.nbc (COUNT)

Michael Stegh r-project at michael-stegh.com
Wed Mar 7 00:07:29 CET 2012


Dear List,

I normally try to solve this kind of problems by myself but I am running out of time and have 
no idea where to start. So I hope you might excuse this rather basic question.

I'm trying to fit a Canonical Negative Binomial model using the ml.nbc function from the 
COUNT package. This function only outputs a "normal" regression table with the parameter 
estimators and so on. But for further testing I need the log-likelihood. And I have no idea how 
to easily obtain it. I think this question extends to all ml-estimators which are "non-standard-
functions". 

Any pointers for me how I can progress?

All the best.

Michael Stegh



More information about the R-help mailing list