Date:         Thu, 1 Jul 2004 18:00:22 -0500
Reply-To:     Paul R Swank <Paul.R.Swank@UTH.TMC.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Paul R Swank <Paul.R.Swank@UTH.TMC.EDU>
Subject:      Re: Getting output from NLMIXED
Comments: To: Jeffrey Stratford <stratja@AUBURN.EDU>
In-Reply-To:  <s0e43653.020@TMIA1.AUBURN.EDU>
Content-Type: text/plain; charset="us-ascii"

Try the Predict statement.

Paul R. Swank, Ph.D.
Professor, Developmental Pediatrics
Medical School
UT Health Science Center at Houston


-----Original Message-----
From: SAS(r) Discussion [mailto:SAS-L@LISTSERV.UGA.EDU] On Behalf Of Jeffrey
Stratford
Sent: Thursday, July 01, 2004 3:05 PM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Getting output from NLMIXED


I'm finally figuring out what I'm doing with the ZIP models and the last
thing I need to do is get a table with predicted values for the dependent
variable.  NLMIXED doesn't appear to like an OUTPUT statement.  Could
someone tell me how to get a table with both the original values and
predicted values?

Thanks,

Jeff

Below is  the SAS code for the NLMIXED procedure

proc nlmixed data=inbu;
 parameters a0=0 to 1 by .2 a1=0 to 1 by .2  b0=0 to 1 by .2 b1=0 to 1 by
.2;  linpinfl = a0 + a1*TK;  infprob  = 1/(1+exp(-linpinfl));
lambda   = exp(b0 + b1*TK);
 if round=0 then
      ll = log(infprob + (1-infprob)*exp(-lambda));
 else ll = log((1-infprob)) + density*log(lambda) - lgamma(density+1) -
lambda;  model density ~ general(ll); output out=temp; run;

****************************************
Jeff Stratford
331 Funchess Hall
Department of Biological Sciences
Auburn University
Auburn, AL 36849
FAX 334-844-9234
http://www.auburn.edu/~stratja
****************************************