[R] lme4 mcmcsamp matrix not PD

Martin Henry H. Stevens HStevens at MUOhio.edu
Thu Dec 21 22:22:49 CET 2006


Hi folks,
A follow up - this problem happens only stochastically. Further, a  
little trial and error indicates that using mathod="Laplace" or  
method="AGQ" reduces the frequency of the problem. Why? I have no  
idea, and even this could be my own wishful interpretation of my  
trial and error.
-Hank
On Dec 20, 2006, at 4:20 PM, Martin Henry H. Stevens wrote:

> Hi folks,
> I am not providing a small replicable example, because I assume the
> problem is related to my quirky data (~650 obs.).
> I am using the latest lme4, matrix and coda, and R 2.4.0.
>
> I frequently get the following error message for this particular lmer
> model. I do not get this message for the vast majority of my uses of
> mcmcsamp.
>
> I was wondering if it might be characteristic of particular kinds of
> problems.
>
>> modb <- lmer(log(basal+1) ~ nutrient*amd +  (1|rack) +  (1|
> gen:amd), data=dat.b2)
>
>> mod.mc <- mcmcsamp(modb, n=10^4)
>
> Error: Matrix is not pd after safe_pd_matrix!
> Error in t(.Call(mer_MCMCsamp, object, saveb, n, trans, verbose)) :
> 	error in evaluating the argument 'x' in selecting a method for
> function 't'
>
> Cheers,
> Hank
>
> Dr. Hank Stevens, Assistant Professor
> 338 Pearson Hall
> Botany Department
> Miami University
> Oxford, OH 45056
>
> Office: (513) 529-4206
> Lab: (513) 529-4262
> FAX: (513) 529-4243
> http://www.cas.muohio.edu/~stevenmh/
> http://www.muohio.edu/ecology/
> http://www.muohio.edu/botany/
>
> "E Pluribus Unum"
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting- 
> guide.html
> and provide commented, minimal, self-contained, reproducible code.



Dr. Hank Stevens, Assistant Professor
338 Pearson Hall
Botany Department
Miami University
Oxford, OH 45056

Office: (513) 529-4206
Lab: (513) 529-4262
FAX: (513) 529-4243
http://www.cas.muohio.edu/~stevenmh/
http://www.muohio.edu/ecology/
http://www.muohio.edu/botany/

"E Pluribus Unum"



More information about the R-help mailing list