[R] PROC MIXED user trying to use (n)lme...

Søren Højsgaard sorenh at mail.dk
Wed Oct 10 00:20:45 CEST 2001

Dear R-users

Coming from a proc mixed (SAS) background I am trying to get into
the use of (n)lme.

In this connection, I have some (presumably stupid) questions
which I am sure someone out there can answer:

1) With proc mixed it is easy to get a hold on the estimated
variance parameters as they can be put out into a SAS data set.
How do I do the same with lme-objects? For example, I can see the
estimated variance matrix when typing the name of the lme-object,
but that does not give me a "handle" on the matrix.

2) I am trying to fit a mixed model, say

    y_{ij} = X_{ij}\beta + U_i + e_{ij}

where Y_{ij}, U_i and e_{ij} all are D-dimensional, and U_i ~
N_D(0,G), and e_{ij} ~ N(0,R). This is easy to do in proc mixed by
including a factor keeping track of how the elements of y_{ij} are
"stacked" on top of each other. This factor is then used in the
model, the random and the repeated statements. In the book by
Pinheiro and Bates (p.28 ff) there is a similar example. Yet, I
don't quite get the "generality" of the example. Can anyone help
with that?

3) With proc mixed it is easy to specify that some covariance
parameters are known (fixed) in advance. How can I do the same
with lme? (To be specific, I would like to specify that the
elements of R in the example above are known).

Thanks in advance

Søren Højsgaard
sorenh at agrsci.dk    http://www.jbs.agrsci.dk/~sorenh
Time flies like an arrow, fruit flies like a banana.

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

More information about the R-help mailing list