[R] fast optimization routines in R

Ravi Varadhan rvaradhan at jhmi.edu
Tue Feb 8 00:27:19 CET 2011


Try the "optimx" package.  It is ideal for doing comparative performance
evaluations of different optimizers for box-constrained problems.  It
unifies about a dozen algorithms under a single function call that is almost
identical to that of `optim'.  You need to set the control option as
`all.methods=TRUE' to get all the algorithms.

Ravi.

-------------------------------------------------------
Ravi Varadhan, Ph.D.
Assistant Professor,
Division of Geriatric Medicine and Gerontology School of Medicine Johns
Hopkins University

Ph. (410) 502-2619
email: rvaradhan at jhmi.edu


-----Original Message-----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of ppinger
Sent: Monday, February 07, 2011 3:45 PM
To: r-help at r-project.org
Subject: [R] fast optimization routines in R


Dear R help archive group, 
I am looking for a maximization routine that I can use to maximize a large
variety of relatively complex likelihoods. I undertand (from previous posts)
that coding the objective function more efficiently can help. However, the
optimization routine employed seems important too. So far, I have tried the
optimization routines optim, maxlik, trust and nlminb. The latter two are
much faster than the first ones but nevertheless, it seems to me as if these
routines were rather slow, when compared to some of the optimizers in
MATLAB.
Is there any general advice you can give about which optimization routines
in R tend to be particularly fast?
Thank you very much, 
Pia
-- 
View this message in context:
http://r.789695.n4.nabble.com/fast-optimization-routines-in-R-tp3265071p3265
071.html
Sent from the R help mailing list archive at Nabble.com.

______________________________________________
R-help at r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.



More information about the R-help mailing list