[R-sig-ME] MCMCglmm

Tanya Pennell T.Pennell at sussex.ac.uk
Thu Oct 16 11:22:15 CEST 2014


I have a query regarding an MCMCglmm output.

My prior and model is as follows:

prior.model.23<-list(R=list(V=diag(2)/84, nu=0.01),G=list     (  G1=list   (V=diag(2)/84, nu=2, alpha.mu=c(0,0), alpha.V=diag(2)/84)))

model.23 <- MCMCglmm(S_relative_fec ~ sex*rep-1,random=~us(sex):line, rcov=~idh(sex):units, family="gaussian", nitt = 100000, burnin = 30000, thin=30, data = GES23newdata, prior = prior.model.23, verbose = FALSE)

I am looking at sex specific genetic variances and their covariance. I use point estimates of the sex specific variance and covariance to calculate the genetic correlation between the sexes. I have used this model with several different data sets, where the genetic correlation is vastly different between them. The problem is that the HPD interval for each of them is the same every time. I was wondering whether there is something wrong with my prior that would cause this problem to arise?

Many thanks,

	[[alternative HTML version deleted]]

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