Date:         Mon, 23 Feb 2004 16:45:24 -0500
Reply-To:     "Elmaache, Hamani" <Hamani.Elmaache@CCRA-ADRC.GC.CA>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "Elmaache, Hamani" <Hamani.Elmaache@CCRA-ADRC.GC.CA>
Subject:      ZI_NOR  NLMIXED_Problem with 2 Codes
Content-Type: text/plain; charset="iso-8859-1"

Hi there.

I have hard time to understand Code 1 works very well, but NOT Code 2.
For Code 2, I get the following error  message:

ERROR: Quadrature accuracy of 0.000100 could not be achieved with 31 points.
The achieved
       accuracy was 1.000000.

In fact, the CODE 2 is first step to set Zero-infated Normal

Thanks for any help.


/**********    Code 1   ********/
%let data=projline2;
%let depvar=habitud;
proc nlmixed data=&data ;
   parms b0=0.6 b1=1  s2e=14 s2u1=37 ;
   mu =b0+ b1*time ;
   mu =mu+u1 ;
   model &depvar ~ normal(mu,s2e);
   random u1 ~ normal(0,s2u1 ) subject=id ;
run;

/**********    Code 2  ********/

proc nlmixed data=&data ;
   parms  b0=0.6 b1=1 s2e=14 s2u1=37 ;
   mu =b0+ b1*time ;
   mu =mu+u1 ;
 prob = pdf('NORMAL',&depvar,mu,s2e);
loglike = log(prob);
model &depvar ~ general(loglike);
   random u1 ~ normal(0,s2u1 ) subject=id ;
run;