R help - Jun 2007 - using lme on multiple datasets in one shot

Dear list,

I would like to do a huge number of lme's using the same design matrix
(and fixed and random effects). Is it possible to do this efficiently?
Doing otherwise is not an option for my example.

Basically, I am wanting to do the following which is possible using lm:

X <- matrix(rnorm(50),10,5)
Y <- matrix(rnorm(50),10,5)
lm(Y~X)  

with lme. Any suggestions?

Many thanks and best wishes,
Ranjan

On 6/21/07, maitra at iastate.edu <maitra at iastate.edu>
wrote:
> Dear list,
> I would like to do a huge number of lme's using the same design matrix
> (and fixed and random effects). Is it possible to do this efficiently?
> Doing otherwise is not an option for my example.
> Basically, I am wanting to do the following which is possible using lm:
> X <- matrix(rnorm(50),10,5)
> Y <- matrix(rnorm(50),10,5)
> lm(Y~X)
> with lme. Any suggestions?
It would not be easy to do this.  Neither lme nor lmer were designed
to make this easy to do.  There is a better chance of accomplishing
this by creating a custom function based on the current lmer but the
modifications required are non-trivial.

This is a reasonable thing to want to accomplish and I will add it to
the "To Do" list for lmer.  However it is not something I will be able
to get to soon.