Date:         Mon, 29 Apr 1996 12:36:33 GMT
Reply-To:     Terry Hyslop <t_hyslop@LAC.JCI.TJU.EDU>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Terry Hyslop <t_hyslop@LAC.JCI.TJU.EDU>
Organization: Thomas Jefferson University
Subject:      Re: Maximum Likelihood Estimation

jargo@utdallas.edu () wrote:
>Is there a PROC that allows a user to specify a likelihood function and
>gives the MLE parameter estimates?  I thought maybe I could do this with
>PROC NLIN, but after reading the manual, I got the impression that it is not
>really set up to do this, although perhaps there is a workaround.
>
>--
>-- Paul A. Jargowsky (jargo @ utdallas.edu)
 
Paul,
 
Proc NLP(Non Linear Programming) is the procedure for your situation.  Call SAS
Tech
Support to get the documentation.  It is available as of version 6.08.  It
offers
many numerical methods, including one which doesn't require derivatives.  It
also
allows the use of linear and/or nonlinear constraints.  If you prefer IML,
there are
also IML routines which have been written to perform the same computations as
Proc
NLP.  This documentation is also available through Tech Support.
 
Terry Hyslop
Biostatistics Section
Thomas Jefferson University
Terry.Hyslop@mail.tju.edu