JISCMail - ALLSTAT Archives

If you wish to generalize your findings to all centers, not just the ones you have studied, entering center as a random effect makes good sense.

I don't see why you would want to have an ar(1) covariance matrix, it suggests a repeated measures design, and you have not indicated that you are getting more than one observation per experimental unit. Perhaps you need to better explain your study deign if you want more complete assistance from this listserver.

John David Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)>>> Óscar Martín Camacho<[log in to unmask]> 5/19/2012 8:40 AM >>>

Hello,

I am trying to analyse data from a longitudinal study using PROC MIXED in SAS software.

This is a multicentre trial and therefore, I was thinking to specify the subject variable in the REPEATED statemet with an autoregressive AR(1) covariance matrix and to include a variable City in the RANDOM statement to assess its importance by likelihood ratio.

I would be grateful for any suggestions related to the covariance matrix and whether including City in the RANDOM statement is a good approach or if there are other alternatives.

Thank you and best regards,

Oscar.

This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information. Any unauthorized use, disclosure or distribution is prohibited. If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.