Date:         Fri, 4 Jun 2010 10:33:49 -0400
Reply-To:     Ryan Black <ryan.andrew.black@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ryan Black <ryan.andrew.black@GMAIL.COM>
Subject:      Re: not getting plots in proc mixed results
In-Reply-To:  <773963205B4CAB4689DE565DB49C530C43E34D484F@UMMSMAIL07.ad.umassmed.edu>
Content-Type: text/plain; charset=ISO-8859-1

Lorna,

I notice that you have a random intercept yet no subject variable in your
RANDOM statement. You have also not specified the type of covariance
structure for the RANDOM statement. Are you comfortable with the default?

Ryan

On Fri, Jun 4, 2010 at 10:13 AM, Simon, Lorna <Lorna.Simon@umassmed.edu>wrote:

> I ran a very simple model in proc mixed and asked for plots, but I'm not
> getting them in the html file. Is it because I got the warning that the
> "final hessian is not positive definite"? or am I specifying something
> wrong? I must admit, this is the 1st time I've tried to use ODS to get
> statistical graphs. So maybe I'm doing something wrong there. here is my
> code:
>
> ods graphics on;
> ods html file="z:\irb study\analysis\silent majority paper\output\protocol
> role.html";
> proc mixed covtest noclprint order=internal;
> class siteid panel protocolid;
> model words =primaryreviewer secondaryreviewer tertiaryreviewer/solution
> ddfm=bw;
> random intercept siteid panel protocolid;
> title1 "IRB study";
> title2 "HLM predicting number of words spoken by protocol_role";
> run;
> ods graphics off;
>
> any help would be appreciated.
>