Stat Twitter, if you are familiar with negative binomial regression, would you DM me? I have some questions. Thanks.
Think @gwern played around with it. I've never needed it. Or at least, never thought I needed it...
-
-
I played around with it in the sense that it's one of the supported families in R's 'glm'. Ultimately, using `family='NegBinomial'` is no different than using `family='gaussian'` or `family='poisson'` - just another parameterization of a generalized linear model.
-
NegBin should produce larger std errors than the others. SPSS poisson≠NegBin. Anyway, I have a simple q. The equation predicts LOGe(Y). But LOGe(0) is undefinable. I don't get it. Do you?
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.