[Rd] bounds violations, infinite loops in optim/L-BFGS-B (PR#671)

Prof Brian Ripley Prof Brian Ripley <ripley@stats.ox.ac.uk>
Tue, 3 Oct 2000 16:51:10 +0100 (BST)

> From: ben@zoo.ufl.edu
> Date: Tue, 3 Oct 2000 11:45:37 -0400 (EDT)


>   On the other hand, I might choose to work on adapting a different set of
> code -- the "nlminb" function in Splus5 appears to be based on a different
> set of Netlib FORTRAN code (dmnfb,dmngb,dmnhb) -- don't know the
> particular reasons for using a different algorithm here.

1) The one used in optim is later and supposedly better.

2) I do try not to copy S-PLUS (as distinct from S).

3) There are quite a few problems with nlminb too.

Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-devel-request@stat.math.ethz.ch