[R-sig-ME] Parameter expansion with MCMCglmm

[Jarrod Hadfield]

Hi,

Is the error meassage

Error in priorformat(if (NOpriorG) { :   V is the wrong dimension for  
some prior$G/prior$R elements

or

Error in priorformat(if (NOpriorG) { :  alpha.V is the wrong dimension  
for some prior$G/prior$R elements

?

I think it should be the latter. If so alpha.V should be a 4x4 matrix  
rather than a scalar, and alpha.mu should be a vector of length 4  
rather than a scalar.

Cheers,

Jarrod.




Quoting Nate Fronk <nrf5017 at psu.edu> on Tue, 18 Jun 2013 15:04:30 -0400:

>
>
> Hello all,
> I am trying to analyze a Before-After Control-Impact experiment. However, my
> model is not converging well and I am attempting to use a prior with  
> parameter
> expansion. I'm using count data (birds per point), well pad density per unit
> area, and time (1=before, 2=after).
>
> prior<list(R=list(V=diag(1),nu=0.002),G=list(G1=list(V=diag(1),nu=0.002,alpha.mu=0,alpha.V=625),G2=list(V=diag(1),nu=0.002,alpha.mu=0,alpha.V=625),G3=list(V=diag(4),nu=0.002,alpha.mu=0,alpha.V=625)))
> modelA<-MCMCglmm(Count~Density*Time,random=~ Block + Block:Point  
> +idh(1+Density*Time):Species,family="poisson",pr=TRUE,prior=prior,data=edge,nitt=40000)
> When trying to use the prior I get the following message:
> Error in priorformat(if (NOpriorG) { :   V is the wrong dimension  
> for some prior$G/prior$R elements
> I realize that my V= value may be wrong but I'm unsure about what  
> the value actually should be. Should I be using my covariance  
> matrix?  Any help anyone could offer would be greatly appreciated.
> Thank you,
> Nate Fronk
> Penn State University
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>


-- 
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.