Re: R-beta: repeated measures

Douglas Bates (
25 Feb 1998 08:34:35 -0600

To: Jim Lindsey <>
Subject: Re: R-beta: repeated measures
From: Douglas Bates <>
Date: 25 Feb 1998 08:34:35 -0600
In-Reply-To: Jim Lindsey's message of Wed, 25 Feb 1998 11:48:07 +0100 (MET)

Jim Lindsey <> writes:

> > Also note that Bates & Pinheiro are in the process of porting lme
> > (linear mixed effects models) to R. From what I have seen, this does
> > at least what SAS PROC MIXED does, only better...
> My libraries also do all of SAS Proc Mixed. One difference with
> respect to nlme is that I have not implemented models with more than
> one random effect, and certainly not for nonlinear models i.e just
> random intercept with nonlinear models, but any distribution, not just
> normal. I am waiting impatiently for that aspect of nlme.

We are porting the lme part of the nlme 3.0 library to R.  Right now we
have it working on some examples but other examples are tripping up on 
some inconsistencies in the behaviour of the data.frame function.

The biggest differences between nlme 2.1, the version currently
available for S/S-PLUS, and nlme 3.0, the development version, is the
ability to work with nested random effects and the incorporation of
trellis graphics presentations of the data and of the model fits.  The
graphics part does not work in R at present.  There are many, many
internal differences between 2.1 and 3.0 in the organization of the
code and the algorithms but they are not as visible at the user level.

I am not all that familiar with SAS PROC MIXED but I am starting to
read about it.  I believe it is not too much of an exaggeration to say
that all the models fit by PROC MIXED can be fit in the representation
we use.

Regarding nlme, there are no facilities for fitting nonlinear
regression models in R at present except by re-expressing the model in
a form suitable for nlm.  Incorporating nonlinear regression into R
would be non-trivial.  I don't expect that Jose' or I will have the
time to work on that in the near future.  Too many other projects in
the hopper, I'm afraid.
r-help mailing list -- Read
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: