Date:         Tue, 7 Jul 2009 09:57:26 -0700
Reply-To:     Steve Denham <stevedrd@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Steve Denham <stevedrd@YAHOO.COM>
Subject:      Re: Proc An infinite likelihood is assumed in iteration 0 because
              of              a nonpositive definite estimated R matrix for
              Subject 16
Comments: To: "cat.." <cat.b41@GMAIL.COM>
In-Reply-To:  <3e6b244e-2c5a-4fa7-8d5f-0ad8bedb5b11@k19g2000yqn.googlegroups.com>
Content-Type: text/plain; charset=iso-8859-1

Catherine,

The most likely reason is that subject 16 has a duplicate record (duplicate in the fixed effects).  Do a quick check (proc means or freq work well) and see if that isn't the case.
 Steve Denham
Associate Director, Biostatistics
MPI Research, Inc.



----- Original Message ----
From: cat.. <cat.b41@GMAIL.COM>
To: SAS-L@LISTSERV.UGA.EDU
Sent: Tuesday, July 7, 2009 11:56:35 AM
Subject: Proc An infinite likelihood is assumed in iteration 0 because of a nonpositive definite estimated R matrix for Subject 16

Hi there,

I'm running a proc mixed for a repeated measurements analysis with
fixed effects.

The model could not be fitted and the following note is displayed in
the log: "An infinite likelihood is assumed in iteration 0 because of
a nonpositive definite estimated R matrix for Subject 16".

Any suggestion to circumvent this ?

Thanks !

Catherine.