Dear all,
I have a problem formulating random effects when using the nlme package.
?nlme says that the random effects can be a formula or a list of formulae.
But when trying to do so, I run into trouble.
(R-1.2.2, nlme-3.1-10, Linux)
library(nlme)
example(nlme) # works nice ...
fm1 # the fitted model
### Trying to update with different random effects
fm2<-update(fm1,random=list(Asym~1,R0~1))
# gives a "parse error"
fm2<-update(fm1,random=list(Asym=~1,R0=~1))
# nlme says that random effects are no formulae
In this case its possible of course to say "random = Asym + R0 ~1"
which
works (Well, it doesn't converge, but nlme tries to execute the code).
This formulation is not possible with my own data unfortunately, as the
random effects are (likely to be) different with respect to two factors: I
need to express something like "random=list(b0 = ~1, b1 =
~treatment)".
When leaving away the random effects I get an error message as well, which
is straigthforward from the above, because
nlme<-function(..., random=fixed, ...){lots-of-code}
~~~~~~~~~~~~
and the fixed effects are expressed as (list of) formulae.
I had the same problem with R-1.2.1 and nlme-3.1-9.
Am I missing something?
Thanks in advance ...
Uli Flenker
Institute of Biochemistry
German Sport University Cologne
Carl-Diem-Weg 6
50933 Cologne / Germany
Phone 0049/0221/4982-493
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
