Dear Jose and all the other S+ users,
First I want to thank you and Prof. Brian Ripley for your helpful answers.
I'll described the missing values problem in more details.
I'm modeling longitudinal data where each one of the subjects was measured
at few time points.
Some of the subjects dropout from the study, i.e. that the RESPONSE of these
subjects is missing.
The analysis in proc MIXED is likelihood based ignorable analysis (i.e. we
using all available cases at each time point),
in lme ,if i understood you correctly, if the response is missing then the
observation is omitted from the dataset. This is a complete case analysis.
My question is if there is a way to overcome this problem and to use all
observation at each time point (and in that way the output from proc MIXED
and lme will be the same).
Now, I'm on my way to get the gls function.
Thank you once again for your help, Ziv.
=========================================================
Ziv Shkedy
Biostatistics
Center for Statistics
Limburgs Universitair Centrum
Universitaire Campus, department WNI
B-3590 Diepenbeek, Belgium
Tel: +32-(0)11-26.82.57
e-mail: ziv.shkedy@luc.ac.be
=========================================================
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|