[R] offset glm.nb issues why so unstable?

[D_Tomas]

Hi, 

I am using the glm.nb" model type to fit some count data with TWO offset
variables. 

I have successfully used this approach  to build scores of models for
several datasets but I am having problems with two in particular.  

Depending on the model I try to fit for a given dataset, I obtain the
following errors/warnings: 


Error: NA/NaN/Inf in foreign function call (arg 1)
or
glm.fit: algorithm did not converge 
or
alternation limit reached


When I remove one of those offset variables (the one I am less confident in)
no error/warning seems to occur. 

I have checked for zero values, 1 values (as log is used), nas but
everything looks fine. 

My gut feeling is that one of the offset variables makes the model unstable.
However, the same approach was succesfully used to fit some other models
with different datasets.  


The question is perhaps, can observational data that appears to be
reasonable (at least in comparison to other datasets that i previously used)
make the use of an offset variable so unstable? 


Any thoughts or advice on ANY aspect are much appreciated. 

Many thanks


--
View this message in context: http://r.789695.n4.nabble.com/offset-glm-nb-issues-why-so-unstable-tp4631730.html
Sent from the R help mailing list archive at Nabble.com.