Dear S+ users,
I have a data set consisting of 3 measurements per individual.
Is it plausible to fit a gls model where cov(1,2) and cov(2,3) elements
of the covariance matrix are free to vary in the optimization
procedure, while the cov(1,3) is fixed to zero?
S-PLUS 6.0.2 Release 1, Windows 98.
Thanks in advance for your suggestions,
Dimitris
=====
Dimitris C. Rizopoulos
ERASMUS Graduate Student
International Study Program in Statistics
Katholieke Universiteit Leuven
Belgium
http://www.kuleuven.ac.be/ucs/isps/
____________________________________________________________
Do You Yahoo!?
Αποκτήστε τη δωρεάν @yahoo.gr διεύθυνση σας στο http://www.otenet.gr
|