[R-sig-ME] multi-trait model in MCMCglmm

[Jarrod Hadfield]

Hi Matt,

Invert the relationship matrix, pass it to ginverse, and use random =  
~us(trait):random.factor  where random.factor contains levels in the  
rownames of the inverse relationship matrix. Make sure the  
relationship matrix is positive definite and coerce the inverse to  
sparse format using as(myinverse, "dgCMatrix").

Cheers,

Jarrod

Quoting Matthew Robinson <matthew.r.robinson at sheffield.ac.uk> on Thu,  
24 Jan 2013 13:21:36 +0000:

> Hi,
>
> I was wondering if it was possible run a bivariate version of the  
> following model in MCMCglmm?
> m1<-MCMCglmm(y~1, random=~idv(Z), data=data, prior=prior)
> Here, Z is a model matrix gained from singular value decomposition  
> of a relationship matrix, and I have used idv to fit a common  
> variance to all terms in Z.
> What I want is to estimate the covariance of the Z effects between  
> two traits. Despite numerous attempts and searching I can't code a  
> model that can do it and any help would be greatly appreciated.
> Many thanks in advance.
> Best wishes,
> Matt
> ------------------------------------------------------
> Dr. Matt Robinson
> NERC Research Fellow
> Department of Animal and Plant Science
> University of Sheffield
> Alfred Denny Building, Western Bank
> Sheffield, S10 2TN, UK
>
> matthew.r.robinson at sheffield.ac.uk
>
> tel:  +44 (0)114 222 4707
> fax: +44 (0)114 222 0002
> ------------------------------------------------------
>
>
>
>
> 	[[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.