Hello Ben,<div><br></div><div>First I tried:</div><div>> mm7b=glmmadmb(count~treat+bronum+bromelv+bromhgt+bromleaf+bromh2o+h2opH+(1|tree),data=ecpad,admb.opts=admbControl(shess=FALSE,noinit=FALSE),family="nbinom")</div>
<div><div><br></div><div>and still got:</div><div>Error in glmmadmb(count ~ treat + bronum + bromelv + bromhgt + bromleaf + : </div><div> The function maximizer failed (couldn't find STD file)</div><div>In addition: Warning message:</div>
<div>running command './glmmadmb -maxfn 500 -maxph 5' had status 1 </div><div><br></div><div>Went down to v.0.7 and all seems well again. Thank you so much! It made for a stressful day up to this point.</div><div>
<br></div><div>Thanks again,</div><div><br></div><div>Shawn </div><div><br></div><br><div class="gmail_quote">On Wed, Jan 25, 2012 at 4:34 PM, Ben Bolker <span dir="ltr"><<a href="mailto:bbolker@gmail.com">bbolker@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="HOEnZb"><div class="h5">On 12-01-25 05:05 PM, Shawn McCracken wrote:<br>
> Dear glmmADMB users,<br>
><br>
> Yesterday I was running my full & reduced models with several different<br>
> distribution families and had no problems except for the warning message<br>
> "Estimated covariance matrix may not be positive definite" followed by a<br>
> string of numbers for a few of the distribution famlies. But when I ran the<br>
> summary() command on the model everything seemed fine. I decided I would<br>
> update to the latest version of glmmADMB to see what happens. Ran the same<br>
> models and the warning was gone. Great! I was on Mac OSX 10.6.8, running R<br>
> v.2.14.0 64-bit, and package glmmADMB v.0.6.5 and updated to glmmADMB<br>
> v.0.7.2.5. I ran the following distributions: poisson, zero-inflated<br>
> poisson, negative binomial, zero-inflated negative binomial, and negative<br>
> binomial type 1. Today I come back to continue working on reduced models<br>
> and I get the following error:<br>
> Error in glmmadmb(count ~ treat + bromleaf + (1 | tree), data = ecpad, :<br>
> The function maximizer failed (couldn't find STD file)<br>
> In addition: Warning message:<br>
> running command './glmmadmb -maxfn 500 -maxph 5 -noinit -shess' had status<br>
> 1<br>
><br>
> I have now gone through restarting everything, downgrading glmmADMB to<br>
> v.0.7.2.4 then back to v.0.7.2.5, and upgraded R to v.2.14.1 with no luck.<br>
> I looked for glmmADMB version 0.6.5 but couldn't find it. Just some<br>
> preliminary testing with models that ran fine yesterday it appears<br>
> that negative binomial (family="nbinom") and negative binomial type 1<br>
> (family="nbinom1") are the two that now give the above error. It is strange<br>
> that zero-inflated negative binomial works fine. I have tried to follow the<br>
> previous few posts about dealing with this problem but with no luch and its<br>
> a little unclear to me if they were on a Mac or not for some of those<br>
> remedies.<br>
><br>
> Really loving all the new functionality of glmmADMB so big thanks to the<br>
> developers. Just bummed its not working for me now, particularly negative<br>
> binomial type 1.<br>
><br>
> Where can I get the glmmADMB v. 0.6.5 that was working?<br>
<br>
</div></div> A couple of thoughts:<br>
<br>
* with a little bit of work I can retrieve the relevant version from<br>
the source code repository, build it, and put it up somewhere. In the<br>
meantime you could go to<br>
<a href="http://www.math.mcmaster.ca/~bolker/R/src/contrib/" target="_blank">http://www.math.mcmaster.ca/~bolker/R/src/contrib/</a> to download version<br>
0.7 (then put it somewhere, setwd() to the appropriate directory, and<br>
install.packages("glmmADMB_0.7.tar.gz",repos=NULL,type="source"))<br>
* a lot of the recent changes in ADMB have to do with tweaks to the TPL<br>
(model definition) file and to ADMB itself that are _supposed_ to make<br>
things more stable in general, but *may* tip a marginally stable model<br>
over the line to unstable -- think of it as a perturbation in a<br>
generally good direction but with some random component ...<br>
* try running with admb.opts=admbControl(shess=FALSE,noinit=FALSE) and<br>
see if that helps at all ...<br>
<br>
Ben Bolker<br>
<br>
<br>
_______________________________________________<br>
Users mailing list<br>
<a href="mailto:Users@admb-project.org">Users@admb-project.org</a><br>
<a href="http://lists.admb-project.org/mailman/listinfo/users" target="_blank">http://lists.admb-project.org/mailman/listinfo/users</a><br>
</blockquote></div><br><br clear="all"><div><br></div>-- <br><br><div><br>Shawn F. McCracken<br>Ph.D. Candidate - Aquatic Resources<br>Department of Biology<br>Texas State University<br>601 University Drive<br>San Marcos, Texas 78666<br>
<br><a href="mailto:smccracken@txstate.edu" target="_blank">smccracken@txstate.edu</a><br><a href="http://uweb.txstate.edu/~sm1216/" target="_blank">http://uweb.txstate.edu/~sm1216/</a><br>_________________________________<br>
<br>Executive Director/Research Coordinator<br>The TADPOLE Organization<br> <br><a href="mailto:smccracken@tadpoleorg.org" target="_blank">smccracken@tadpoleorg.org</a><br><a href="http://www.tadpoleorg.org" target="_blank">www.tadpoleorg.org</a><br>
<br>"It is horrifying that we have to fight our own government(s) to save the environment." - Ansel Adams </div><br>
</div>