?? Hello everyone,
?? I am reading a lot of documentation at the moment about prior
specification in MCMCglmm, and there is still something I have not
very clear. According to some sources, there are mainly two elements
to take into account when defining a prior:?? - An R structure that
needs to be specified for each fixed effect. And
?? - A G structure for each random effect.
?? However, in other sources another element, B, is introduced as
well, which refers to fixed effects too. In Jarrod's Course Notes, I
see that both R and B have to do with fixed effects: R specifies V
and nu arguments for the variance, and B specifies V and mu elements
for the mean.
??
?? My confusion comes from the fact that almost every time I see an
example of a prior, it just has two elements, R (for fixed effects)
and G (for random effects). Why is this? Is it not so important to
define a prior for the mean? Is it enough with a prior specification
for the variance of fixed effects and another one for all the random
effects?
?? Thank you very much in advance.?? Iker
?