[R-sig-ME] Bivariate (or multivariate) model in MCMCglmm

[Napoleon Vargas]

Dear everyone,

I've been playing with MCMCglmm for a while, and I've found it very useful.
I wanted to know if its possible to restrict specific residual variances in
a bivariate or multivarate model. The model in question would have one (or
two) Gaussian traits, and one (or two) "Threshold" or Ordinal traits.I
would like to restrict the residual variances for the threshold traits to
1. Would that be feasible (or a sensible thing to do)?.

I'd appreciate any information.


Cheers,

-- 
*Napo Vargas J.*
*Ph.D. Student*
*Breeding and Genetics*
*Department of Animal Science*
*University of Nebraska - Lincoln*

	[[alternative HTML version deleted]]