[R-sig-ME] glmer and poisson distributions and P values

[Luca Borger]

Hi Emma,

(my two cents) It is usually very helpful to do first a search on the mailing list archives. For example, at:


############################################################################
http://markmail.org/message/oogvmvpxldr5i57q

you can find a reply by Prof. Bates:

"In mer_finalize(ans) : false convergence (8)

That's a warning that the parameter estimates are suspect but it's not
an error. I believe that a fitted model would have been returned.

I suggest you turn on the verbose option in the optimizer and examine
the progress of the iterations.  In the original call to glmer you add
verbose = TRUE.  If re-calling "mer_optimize" as above then change the
FALSE to TRUE."
#############################################################################


In addition, more information would be needed about your data etc. For example, in some cases similar convergence problems can be solved by centreing/standardising the predictor variable (but I may get corrected on this by the experts on this list).

HTH



Cheers,

Luca



----- Messaggio originale -----
Da: "Emma Stone" <Emma.Stone at bristol.ac.uk>
A: r-sig-mixed-models at r-project.org
Inviato: Lunedì, 16 marzo 2009 11:21:00 GMT -05:00 U.S.A./Canada, stati orientali
Oggetto: [R-sig-ME] glmer and poisson distributions and P values


Dear all,

I am trying to run a glmer model with count data,with one random factor. 
When I run it with more than one predictor all is well, but when there is
only one predictor in the model I get this warning message, can anyone 
explain this at all?

Warning message: In mer_finalize(ans) : false convergence (8)

Also, for other models in the canidate set I am looking at interactions 
between predictors and I want to get the P values for these interactions,
any ideas how I can do this from a glmer model?

Thanks

Emma


--

_______________________________________________
R-sig-mixed-models at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models