[R-sig-ME] problem WITH mcmcsamp function in lme4 on Mac OSX

José Manuel Blanco Moreno jmblanco at ub.edu
Thu Dec 11 22:42:59 CET 2008

I had the very same problem today on a PC. I have updated lme4 (just in 
case) but it did not solve the problem.
José M.

HENERY Martin escribió:
> I was wondering if anyone else had experienced this problem and maybe  
> could identify it as a possible bug rather than a problem peculiar to  
> my particular computer.
> I have been trying to get confidence intervals via the mcmcsamp  
> function in lme4 package but cannot get it to work on a mac running  
> OS X. i ran exactly the same analyses on a PC with exactly the same  
> commands and data and it worked fine. I removed both the matrix and  
> lme4 packages from the library on the Mac (based on an unrelated  
> problem with the lme4 package that I read about in the mailing list)  
> and reinstalled them but to no avail. When I run the mcmcsamp  
> function I get this error message (sorry its partly in french)
>  > model.glmer4=glmer(leaves.eaten ~ Geocytotype+Cypho.origin + (1| 
> TEST), data = cyphoplantfeeding, family=poisson)
>  > summary(mcmcsamp(model.glmer4 , n = 10000))
> Erreur dans .local(object, n, verbose, ...) : Update not yet written
> Erreur dans summary(mcmcsamp(model.glmer4, n = 10000)) :
>    erreur lors de l'évaluation de l'argument 'object' lors de la  
> sélection d'une méthode pour la fonction 'summary'
> The same script runs fine on a PC so is what is the "update not yet  
> written" referring to?
> Martin Henery
> 	[[alternative HTML version deleted]]
> ------------------------------------------------------------------------
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

José M. Blanco Moreno

Dept. de Biologia Vegetal (Botànica)
Universitat de Barcelona
Av. Diagonal 645 
08028 Barcelona SPAIN

phone: (+34) 934 039 863
fax: (+34) 934 112 842
e-mail: jmblanco at ub.edu 

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