[R-sig-ME] glmmADMB package
bbolker at gmail.com
Sat Jan 15 15:10:07 CET 2011
-----BEGIN PGP SIGNED MESSAGE-----
I'm pretty sure that this is an issue with a file path with spaces in
it. I thought that I had fixed a problem like this previously.
It's not clear from your e-mail below: did you download the zip file
from the Otter Research web page, or from r-forge? If the latter, then
I still have some bug chasing to do.
Can you please start R, load the package (library("glmmADMB")), run
sessionInfo(), and post the results? Also, please post the full
It is possible to use MCMCglmm for this model (or rather a closely
related one that uses the lognormal-Poisson rather than the negative
binomial to model overdispersion), but the syntax turns out to be a bit
trickier than glmmADMB in this case. Let me know if you end up pursuing
that direction, although I'd like to get the glmmADMB problems sorted
out in any case.
On 11-01-14 12:29 PM, Thomas Merkling wrote:
> Thanks for the answers !
> I finally achieved to install the package on R 2.7.2 but not on more
> recent versions.
> did not work , it tells that glmmADMB is not available.
> I dowloaded the .zip file and installed it from R then replacing the two
> files from the both.zip (as indicated on :
> But when I try to fit a model I got this message:
> 'C:/Users/thomas' n'est pas reconnu en tant que commande interne
> ou externe, un programme ex‚cutable ou un fichier de commandes.
> Erreur dans glmm.admb(Nb ~ ClAge * SexChick * EggRank, random = ~Nestcol, :
> The function maximizer failed
> De plus : Warning messages:
> 1: In dir.create(dirname) : '_' existe déjà
> 2: In glmm.admb(Nb ~ ClAge * SexChick * EggRank, random = ~Nestcol, :
> Argument "link" ignored for other familes than "binomial" (exponential
> link used)
> 3: In file.remove(std_file) :
> impossible d'effacer le fichier 'nbmm.std', à cause de 'No such file or
> 4: In shell(paste(.path.package("glmmADMB"), "/admb/", file_name, ".exe", :
> merkling/Documents/R/win-library/2.7/glmmADMB/admb/nbmm.exe -maxfn 500 '
> execution failed with error code 1
> I don't understand why ...
> I heard about MCMCglmm and I will try it.
> Do I need to fit the overdispersed version of the Poisson model with
> another function, lmer for example ?
> And what is it for ?
> Thanks ...
> Le 14/01/2011 17:46, Ben Bolker a écrit :
>> I think it would be better to retrieve the appropriate binary package
>> from<https://r-forge.r-project.org/R/?group_id=847> (or navigate to
>> r-forge, to the glmmADMB project, and then to the "R packages" page; it
>> will probably install just fine if you download the zip file and
>> "install from local packages".
>> By the way, how did you try to download the package, and what was the
>> result? Did
>> not work?
>> If you tried to install from CRAN it wouldn't have worked anyway --
>> glmmADMB is hosted on r-forge rather than CRAN because it includes
>> binaries (which aren't allowed on CRAN).
>> On 11-01-14 08:29 AM, Dennis Murphy wrote:
>>> A simple Google search on 'glmmADMB R' returned this as the first hit:
>>> Perhaps that will help...
>>> On Fri, Jan 14, 2011 at 4:15 AM, Thomas
>>> Merkling<Thomas.Merkling at cict.fr>wrote:
>>>> Dear list members,
>>>> I want to fit Zero inflated Poisson mixed models and for that I
>>>> expected to
>>>> use the glmmADMB package which seemed to do that. However I was not
>>>> able to
>>>> download the package. I'm using R 2.11.1. For which R version has the
>>>> package been developed ?
>>>> Is there another package authorizing to use to this type of model ?
>>>> Thanks for your help.
>>>> Thomas Merkling
>>>> R-sig-mixed-models at r-project.org mailing list
>>> [[alternative HTML version deleted]]
>>> R-sig-mixed-models at r-project.org mailing list
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
-----END PGP SIGNATURE-----
More information about the R-sig-mixed-models