repeated measures

[<tmhack@fuse.net>]

Hi All,

I am using S-Plus 2000 Profession Release 2 on Windows XP.
I want to perform a regression of repeated measures data.  I have observations 
y after treatment x for several individuals.  
Each individual is tested 4 times at each level of x.  
My model is y ~ constant1 - beta1*(beta2 - x). 
My question is how to set up the model for nlme.  I 
worked for some time and arrived at the following:

>gd <- groupedData(y ~ x | Subject, data=d, outer=~Sex)
>nlme(model=y~11.37-beta1*(beta2-x), fixed = beta1+beta2~1, 
     data=gd, start=c(0.1, 100))

This converges to a solution, but I am still not confident that 
it is doing what I intended.  That is, perform the regression 
while accounting for the fact that some of the data are from 
the same individual.  Can anyone tell if I have set this up 
correctly?

Thanks,
Eric