[R-sig-ME] brms: Multivariate LMEM with systematic missing data

[Jarrod Hadfield]

Hi,

I'm just about to board a plane, so briefly, the equivalent MCMCglmm 
code is:

m1 <- MCMCglmm(cbind(dep1, dep2, dep3) ~ trait-1+trait:group, random = ~us(at.level(trait,1):(1+meas)):id+us(at.level(trait,2):(1+meas)):id+us(at.level(trait,3):(1+meas)):id,
                rcov = ~us(trait):units, data = dat, family = c("gaussian",
"gaussian", "gaussian"),
                     verbose = FALSE)
  
Cheers,

Jarrod


On 02/02/2017 10:04, Koen Neijenhuijs wrote:
> m1 <- MCMCglmm(cbind(dep1, dep2, dep3) ~ group-1, random = ~us(1+meas):id,
>                 rcov = ~us(trait):units, data = dat, family = c("gaussian",
> "gaussian", "gaussian"),
>                      verbose = FALSE)


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