Message-ID: <48056BE0.1000104@pdf.com>
Date: 2008-04-16T03:00:48Z
From: Spencer Graves
Subject: prediction intervals from a mixed-effects models?
In-Reply-To: <loom.20080414T171512-999@post.gmane.org>
Dear Gregor:
Thanks for the suggestion. Scanning "help(pac=Zelig)", I didn't
see anything that sounded to me like "prediction intervals from a
mixed-effects model". Without a more focused suggestions, it seems
simpler to me to transform somehow the MCMC results into the desired
prediction intervals.
Thanks again.
Spencer
p.s. I noticed that I had inadvertently not copied into my email an
line needed to make my example self contained. The following should
work, as far as it goes:
library(lme4)
library(MEMSS)
set.seed(3)
fm3r <- lmer(distance~age*Sex+(1|Subject), Orthodont)
samp3r <- mcmcsamp(fm3r, n=10000)
Gregor Gorjanc wrote:
> Take a look at Zelig. There is a function that worked with previous versions of
> lme4.
>
> Gregor
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
>