Date:         Wed, 11 Apr 2007 13:42:42 -0400
Reply-To:     Jerry Davis <jwd@UGA.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Jerry Davis <jwd@UGA.EDU>
Subject:      Re: Proc Glimmix without Random _residual_ VS Glimmix Macro
In-Reply-To:  <BAY139-F32F5170940E16FAAADF09FC15F0@phx.gbl>
Content-Type: text/plain; charset="US-ASCII"

Anne-Elie carsin wrote:

> proc glimmix DATA = may06;
>      class surgeon hospital hosp_ter;
>      id surgeon hospital ;
>      model postop=hosp_ter / ddfm=kr oddsratio link=logit dist=binomial ;
>      random int  / subject=hospital;
>      random int  / subject=surgeon(hospital) ;
>      nloptions tech=nrridg;
>      random _residual_;
> run;
>
> I just don't get why the RANDOM _residual_; is needed in the Proc Glimmix?
> Results are slighlty different when omitted and I wonder what results would
> be the correct ones? Any idea?


It isn't necessary and you shouldn't need it for your model.  RANDOM
_residual_ serves the same purpose in GLIMMIX as the REPEATED statement
does in MIXED.  So, if you don't need to model repeated effects, leave
it out.

It can probably used for other nifty things, but its purpose is to model
R side effects.

Jerry
--
Jerry Davis
Experimental Statistics
UGA, CAES, Griffin Campus