topic marquardt method in Statistical Procedures https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207591#M11161 <HTML><HEAD></HEAD><BODY><P>Can anybody provide me the complete proc of marquardt method?</P></BODY></HTML> Tue, 16 Jun 2015 16:45:10 GMT afgdurrani0 2015-06-16T16:45:10Z marquardt method https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207591#M11161 <HTML><HEAD></HEAD><BODY><P>Can anybody provide me the complete proc of marquardt method?</P></BODY></HTML> Tue, 16 Jun 2015 16:45:10 GMT https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207591#M11161 afgdurrani0 2015-06-16T16:45:10Z Re: marquardt method https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207592#M11162 <HTML><HEAD></HEAD><BODY><P>Without further details, I can guess only that you're looking at nonlinear regression (PROC NLIN) with METHOD=<SPAN style="font-size: 13px; background-color: #ffffff; font-family: 'Helvetica Neue', Helvetica, Arial, 'Lucida Grande', sans-serif;">MARQUARDT for the fitting algorithm.&nbsp; This community can provide much more help if you provide more details on the problem.&nbsp; There is some<A href="http://www.ats.ucla.edu/stat/sas/library/SASNLin_os.htm"> basic guidance here in this doc from UCLA</A>.&nbsp; And some <A href="http://support.sas.com/rnd/app/ets/papers/nonlinearmodels.pdf">general overview of related procedures here</A>.</SPAN></P><P><SPAN style="font-size: 13px; background-color: #ffffff; font-family: 'Helvetica Neue', Helvetica, Arial, 'Lucida Grande', sans-serif;"><BR /></SPAN></P><P><SPAN style="font-size: 13px; background-color: #ffffff; font-family: 'Helvetica Neue', Helvetica, Arial, 'Lucida Grande', sans-serif;">Chris<BR /></SPAN></P></BODY></HTML> Wed, 17 Jun 2015 01:02:44 GMT https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207592#M11162 ChrisHemedinger 2015-06-17T01:02:44Z Re: marquardt method https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207593#M11163 <HTML><HEAD></HEAD><BODY><P>The Marquardt least-squares optiomization method is available by using the NLPLM subroutine in SAS/IML.&nbsp; The <A href="http://support.sas.com/documentation/cdl/en/imlug/66845/HTML/default/viewer.htm#imlug_langref_sect270.htm">documentation of the NLPLM subroutine</A> describes the algorithm as "an efficient modification of the trust-region method."</P></BODY></HTML> Wed, 17 Jun 2015 19:42:35 GMT https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207593#M11163 Rick_SAS 2015-06-17T19:42:35Z Re: marquardt method https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207594#M11164 <HTML><HEAD></HEAD><BODY><P>Dearest sir,</P><P>Thank you for your kind reply.</P><P>I am a fisheries specialist and frequently use SAS for my statistics problems. In a research paper, I have studied about the modified form of Marquardt method. I want you to have look at this modified method (below link) and kindly provide me your comments and view over this modified form of Marquardt method ; that is it correct? Thanks alot</P><P></P><P><A href="https://drive.google.com/file/d/0B6WGzXbl-1cQb2RWMmlPVU4zZzQ/view" title="https://drive.google.com/file/d/0B6WGzXbl-1cQb2RWMmlPVU4zZzQ/view">Program Summary - general model.pdf - Google Drive</A></P></BODY></HTML> Wed, 01 Jul 2015 08:24:33 GMT https://communities.sas.com/t5/Statistical-Procedures/marquardt-method/m-p/207594#M11164 afgdurrani0 2015-07-01T08:24:33Z