|Date: ||Thu, 15 Nov 2001 14:10:06 -0500|
|Reply-To: ||"Muller, Denis (NIA/IRP)" <MullerD@GRC.NIA.NIH.GOV>|
|Sender: ||"SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>|
|From: ||"Muller, Denis (NIA/IRP)" <MullerD@GRC.NIA.NIH.GOV>|
|Subject: ||Doubly repeated measures in Proc Mixed|
|Content-Type: ||text/plain; charset="ISO-8859-1"|
> Hello. I'm trying to look at the pre-post drug administration effect on
> the time course of c-peptide levels during an experimental procedure. I'd
> like to set up a simple mixed effects model (oxymoron) using the following
> parameters : CP= c-peptide conc., TIME = time of CP measurement
> (0,1,2,3,4,5,6,7,8,9), EXP=Pre or Post classification, ID=subject id.
> There are nine subjects in the dataset.
> When I use the following syntax -
> proc mixed;
> class id time exp;
> model cp=exp time exp*time;
> repeated time exp/sub=id type=cs;
> I get the error -
> ERROR: Only one REPEATED effect is allowed with this covariance structure.
> I've tried other covariance structures (UN, Ar(1))and have gotten the same
> Any suggestions or useful references appreciated. Thanks in advance.
> Denis C. Muller
> National Institute on Aging
> Baltimore, MD 21224-2735