[R-sig-ME] glmmadmb - problems understanding the output when run in cmd console

Diego Pavon diego.pavonjordan at gmail.com
Tue Jan 12 22:31:31 CET 2016

Hello all

I am trying to run GLMM, ZIP and ZINB using glmmadmb package. I have count
data and I want to see whether there is an effect of temperature on the
counts of 12 species over a period of 29 years (counts in ca. 2000 sites).
I am having troubles getting the models converged when running them in
RStudio. For some reason, they do run nicely when I use the windows command
line (cmd).
Then I get all these files with the results (.eval, .par, .std, etc...).

My question is, how do I read those? I can open them in R or a text editor,
but I can't understand what is what. In the glmmadmb website there is a
small description but I can't really understand. for instance, the betas
that I get in the .par file are not the actual estimates. In the website,
It says:

"BETA: fixed-effect parameter estimates: note that these are the versions
of the parameters fitted internally, using an orthogonalized version of the
original design matrix, not the original coefficients (if this means
nothing to you, as it might well, just accept that these are transformed
versions of the parameters)."

So, could someone help me understanding these "results files" that I get
when running glmmadmb outside R, and how to get the estimates, SE and
significance of the covariates?

Thank you very much.


*Diego Pavón Jordán*

*Finnish Museum of Natural History*
*PO BOX 17 *

*Helsinki. Finland*


	[[alternative HTML version deleted]]

More information about the R-sig-mixed-models mailing list