LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous (more recent) messageNext (less recent) messagePrevious (more recent) in topicNext (less recent) in topicPrevious (more recent) by same authorNext (less recent) by same authorPrevious page (May 2003, week 1)Back to main SAS-L pageJoin or leave SAS-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Fri, 2 May 2003 09:31:33 -0700
Reply-To:     Dale McLerran <stringplayer_2@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Dale McLerran <stringplayer_2@YAHOO.COM>
Subject:      Re: MLE regression with hill quadratic climbing
In-Reply-To:  <1A399BD8E0E4D511B5000020ED069A3A0115D96A@msexchange.iso-ne.com>
Content-Type: text/plain; charset=us-ascii

Speaking of IML, it has several modules implementing nonlinear optimization techniques. IML has the following methods: conjugate gradient, double dogleg, Nelder-Mead simplex, Newton-Raphson, Newton-Raphson ridge, quasi-Newton, quadratic, and trust-region optimization methods. Perhaps the quadratic optimization method is the optimization method that is sought. SAS OR has all these optimization methods an more. (See PROC NLP documentation.)

In addition to PROC NLIN, the procedure NLMIXED offers a wide variety of optimization techniques: virtually all of the optimization techniques available in IML are available in the procedure NLMIXED. In my opinion, the NLMIXED procedure has the advantage of ease of specification of the likelihood function.

Dale

--- "Manolakos, Peter" <pmanolakos@ISO-NE.COM> wrote: > Stuart, > > I am not familiar with this optimization algorithm, of course, that > fact does not imply it is not utilized by SAS. Consider PROC NLIN > for instance. According to the SAS documentation there are four > computational methods used to derive estimates (all of which are > basically standard optimization algorithms): steepest descent, > Gauss-Newton, Newton, Marquardt. > > Now, one could code easily the objective function in IML. And then > you could code that method too. > > Perhaps if you provide the list with additional information someone > could answer your question. > > -----Original Message----- > From: Stuart Dennon [mailto:sdennon@HOTMAIL.COM] > Sent: Thursday, May 01, 2003 10:20 PM > To: SAS-L@LISTSERV.VT.EDU > Subject: MLE regression with hill quadratic climbing > > > The following question was asked by a member of the Australian > Academic > SAS User Group - with no answer as yet. > > Dear All, > > Has anyone here ever used SAS to do the maximum likelihood estimation > (by using the quadratic hill-climbing algorithm)? > > What SAS proc can be used? > > Cheers, > > > Any ideas or assistance would be kindly appreciated. > > Thanks in anticipation > > Stuart Dennon

===== --------------------------------------- Dale McLerran Fred Hutchinson Cancer Research Center mailto: dmclerra@fhcrc.org Ph: (206) 667-2926 Fax: (206) 667-5977 ---------------------------------------

__________________________________ Do you Yahoo!? The New Yahoo! Search - Faster. Easier. Bingo. http://search.yahoo.com


Back to: Top of message | Previous page | Main SAS-L page