Hi all,
I've searched threads and read up on some ways of doing this but I'm having
a hard time to get it to work. Here's my basic problem. I have the
following linear mixed model
y = Xb+Zu+e
where u~N(0,s^2*K) where K is a matrix.
I read a thread that basically suggested to decompose Zu into ZPD^(1/2)
D^(-1/2)P'u so that (D^(-1/2)P'u)~N(0,s'^2I) but I'm not sure how to specify
the new ZPD^(1/2) incidence matrix using the lme and lmer packages. The
closest I got was with the lmer package, where I tried using the doFit=FALSE
flag and manually setting the 'Zt' matrices but that doesn't seem to work.
With the lme package, I'm not sure if it's possible to create a new pdMat
class or if I can manually set the Z design matrix. I tried using the
model.matrix commands but that only reads the current design. Thanks in
advance to whoever can help out.
~Jimmie
[[alternative HTML version deleted]]
______________________________________________
[email protected] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
