Groups > comp.soft-sys.math.mathematica > #16591

Re: Some linear constraints seem to be ignored in function

Path	csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!news.glorb.com!newspump.sol.net!posts.news.twtelecom.net!nnrp3.twtelecom.net!not-for-mail
From	Daniel Stariolo <daniel.stariolo@gmail.com>
Newsgroups	comp.soft-sys.math.mathematica
Subject	Re: Some linear constraints seem to be ignored in function
Date	Sun, 2 Mar 2014 06:26:22 +0000 (UTC)
Sender	steve@smc.vnet.net
Approved	Steven M. Christensen <steve@smc.vnet.net>, Moderator
Message-ID	<leuiue$sd7$1@smc.vnet.net> (permalink)
References	<20140222080533.BDB3B6A13@smc.vnet.net> <lec8vh$l39$1@smc.vnet.net>
Lines	4
Organization	Time-Warner Telecom
NNTP-Posting-Date	02 Mar 2014 06:31:57 GMT
NNTP-Posting-Host	00a2eed5.news.twtelecom.net
X-Trace	DXC=_UZB9AJ5h9AVS1USibAf`FC_A=>8kQj6M;[h;PUXBgbDHnHVmV3ecQDEFiONJ7[GoF;S18BA26oSM
X-Complaints-To	abuse@twtelecom.net
Xref	csiph.com comp.soft-sys.math.mathematica:16591

Show key headers only | View raw

Bob, thank you very much for your help. Indeed, changing the Method allows to get some improvement. Nevertheless stochastic methods like DifferentialEvolution or SimulatedAnnealing make a random walk on parameter space and when the excursion goes out of bounds the search stops with the message on the complex value of the logarithm. Then, you get a poor result, generally even far from a local minimum.

A satisfactory workaround for the out of bounds problem that seems to work very well was proposed in this link "http://mathematica.stackexchange.com/questions/42999/nminimize-go-out-of-bounds-on-some-linear-constraints" by Daniel Lichtblau.

Back to comp.soft-sys.math.mathematica | Previous | Next — Previous in thread | Find similar

Thread

Re: Some linear constraints seem to be ignored in function Bob Hanlon <hanlonr357@gmail.com> - 2014-02-23 07:45 +0000
  Re: Some linear constraints seem to be ignored in function Daniel Stariolo <daniel.stariolo@gmail.com> - 2014-03-02 06:26 +0000

csiph-web