R help - Nov 2011 - predict lmer

Dear all,

I've been reading for many days trying to predict with lmer but I
haven't
managed to do it.
I've fitted an allometric model for trees where I have included climatic
variables and diameter in the fixed part and
in the random part I've included the experimental sites where trees are and
also their provenance region.
The model is like this :

f431<-lmer(log(H05)~log(DN05)*(PwS+ PoS+ PpS+
TpS)+PoP:log(DN05)+PwP:log(DN05)+(log(DN05)-1|P)+(log(DN05)-1|S/B)+(log(DN05)|SP),
data=data)

once I 've  fitted the model  I would like to make predictions for a new
data frame of values.
But i don't know how to do it.
For example I am interested in know how height varies (at a fixed diameter)
for a mean rainfall of a site in each provenance region to contrast with
how height response is for a mean rainfall provenance of each site and
things like this.

If someone can give any suggestion I will be pleased to listen to.

Thanks very much for your time,

The best

Natalia

	[[alternative HTML version deleted]]

Natalia Vizca?no Palomar <natalia.vizcaino.palomar <at> gmail.com>
writes:
> I've been reading for many days trying to predict with lmer but I
haven't
> managed to do it.
> I've fitted an allometric model for trees where I have included
climatic
> variables and diameter in the fixed part and
> in the random part I've included the experimental sites where trees are
and
> also their provenance region.
  [snip]
> f431<-lmer(log(H05)~log(DN05)*(PwS+ PoS+ PpS+
> TpS)+PoP:log(DN05)+PwP:log(DN05)+(log(DN05)-1|P)+
    (log(DN05)-1|S/B)+(log(DN05)|SP),
> data=data)
 
  [snip snip]

  Have you looked at the code on http://glmm.wikidot.com/faq about this ... ?
If you look at it and still can't work it out I would suggest that
requesting
help on the r-sig-mixed-models list will be more useful ...

  Ben Bolker