[R-sig-ME] STD file not found error with glmmADMB

Ben Bolker bbolker at gmail.com
Mon Jun 25 10:08:05 CEST 2012


Joshua Wiley <jwiley.psych at ...> writes:

> 
> Hi Melanie,
> 
> This bit, "function maximizer failed" suggests that it is having
> difficulty estimating your model.  What happens if you try a simpler
> model, (the simplest being intercept only)?  My memory is vague, but
> it seems like Ben Bolker may have said at one point that using
> variables from the global environment was not tested much in
> glmmadmb(), so you might also try using the data argument.  Which
> segues nicely into my last point, which is that it would be helpful to
> at least show us the output from
> 
> str(yourdataset)
> 
> so that we know the size, types of variables, etc.
> 
> Cheers,
> 
> Josh
> 
> On Sun, Jun 24, 2012 at 4:57 AM, Melanie Browne
> <meluchbrowne <at> hotmail.com> wrote:

> > Hello, I am using the glmmADMB package (I downloaded it using:
> install.packages("glmmADMB", repos="http://r-forge.r-project.org"),
> I am not sure which version it is). I'm on R version 2.15.1 GUI 1.52
> Leopard build 64-bit, and yes, I'm on a mac.  

> I'm trying to run
> the following function: fit= glmmadmb(fruit ~
> surroundings*town+(1|site), family='nbinom1', zeroInflation=TRUE) >
> and got the following error: dyld: unknown required load command
> 0x80000022Error in glmmadmb(fruit ~ surroundings * town + (1 |
> site), family = "nbinom1",  :   The function maximizer failed
> (couldn't find STD file) > My variables are fruit counts and there
> is overdispersion.  

  What MacOS version are you running?  I strongly suspect it's
MacOS 10.5 rather than 10.6 ... we've had problems with this in
the past (see the discussion at http://glmmadmb.r-forge.r-project.org/ )

  If you google 'glmmadmb "unknown required load command"' you get

https://stat.ethz.ch/pipermail/r-sig-mixed-models/2011q4/007155.html

  I added code to glmmADMB that I thought would detect this
situation, but it may not have worked.  Could you please
run the following commands and post the results?

library(glmmADMB)
sessionInfo()
R.version$os
system("uname -v")



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