Hello everyone,
I have a dataset with repeated measures data on
several subjects which are inter-correlated. Thinking
that clustering is the way to handle the correlation
I've tried to use the 'cluster' argument in my model
as I've seen it in an old example. But, I'm getting
the following error when I'm running the model using
nlme.
Problem in nlme: argument cluster= not matched:
nlme(BodyWeight ~ a + (b1 * Day) + (b2 * Day^2),
fixed = a + b1 + b2 ~ 1,
start = c(5.3, 0.1,0.1),
method = "REML",
cluster = cage,
data = growth, na.action = na.omit)
I've seen that the current 'nlme' function doesn't
have this argument.
Is there another way to handle the correlation.
Any help on this topic/problem is greatly appreciated.
Thank you in advance,
Savi
Statistician, FDA - Jefferson Labs
_______________________________
Do you Yahoo!?
Declare Yourself - Register online to vote today!
http://vote.yahoo.com
|