Skip to content

Define prior in MCMCglmm

1 message · Maria Paola Bissiri

Original
Maria Paola Bissiri
# 
Dear list members,
how should I define the prior for my MCMCglmm model with random =  
~us(1+subj_ID):fin_B?
fin_B is dichotomous, subj_ID has 86 levels (the experiment participants).
The response variable resp_X ist dichotomous.

When I run for instance the following ...
(an attempt inspired by  
http://hlplab.wordpress.com/2009/05/07/multinomial-random-effects-models-in-r/)

k <- length(levels(fallmid$resp_X))
I <- diag(k-1)
J <- matrix(rep(1, (k-1)^2), c(k-1, k-1))
prior = list(R = list(fix=1, V=0.5 * (I + J), n = 4),
              G = list(
                  G1 = list(V = diag(1), n = 2),
                  G2 = list(V = diag(1), n = 2)))

fallmid.MCMCglmm <- MCMCglmm(resp_X ~ lang * ini_pch + lang * manner +  
lang * fin_B,
                              random = ~us(1+subj_ID):fin_B,  
family="categorical", data=fallmid,
                              prior=prior
                              )

.... I get the following error message.

Error in MCMCglmm(resp_X ~ lang * ini_pch + lang * manner + lang * fin_B,  :
   prior$G has the wrong number of structures

How should I define G? Is R fine as it is?
I researched everywhere in Internet and asked people about it, but I  
don't understand how V and n should be indicated. I would be grateful  
if you could help.

Kind regards,
Maria Paola