Re: [R] modeling language for optimization problems

From: roger bos <roger.bos_at_gmail.com>
Date: Thu 06 Oct 2005 - 22:34:50 EST

Prof,
 My apologies if you believe I was contradicting you. I though I was concurring with you that 1) R does not have anything that is as powerful as NuOpt and 2) there is no 'one' optimization package in R. I did, however, want to point out that R does have separate packages that offer a fairy broad range of the simpler types of optimzation. I mentioned that R can do QPs , LPs, and integer programming, among others.

Thanks,
 Roger
 On 10/3/05, Prof Brian Ripley <ripley@stats.ox.ac.uk> wrote:
>
> On Mon, 3 Oct 2005, roger bos wrote:
>
> > As others have alluded, R does not have any one package that is as
> > versatile and powerful as NuOpt, but R does have many different
> optimization
> > packages, so you can do LP, QP, Integer programming, and many more types
> of
> > optimization, all without having to learn a new language. But being
> free,
> > they are admittedly not as powerful as NuOpt. The main thing I know you
> can
> > do in NuOpt that you cannot do in R (to my knowledge) is mixed integer
> > quadratic programming, which would be nice, but there are many work
> arounds.
>
> Please tell us which R packages provide
>
> >> - primal-dual interior point method based on line search for general
> >> Convex Programming (CP) models including convex Quadratic Programming
> >> (CQP) models.
> >>
> >> - primal-dual interior point method based on trust region method for
> >> general Non-Linear Programming (NLP) models.
> >>
> >> - primal-dual interior point method based on quasi-Newton method for
> >> general Non-Linear Programming (NLP) models.
> >>
> >> - active set method for convex Quadratic Programming (CQP) models and
> >> mixed integer Quadratic Programming(MIQP) models.'
>
> and how you found them? I get
>
> > help.search("interior point")
> ...
> rq.fit.fn(quantreg) Quantile Regression Fitting via Interior Point
> Methods
> rq.fit.fnb(quantreg) Quantile Regression Fitting via Interior Point
> Methods
> rq.fit.fnc(quantreg) Quantile Regression Fitting via Interior Point
> Methods
> > help.search("convex programming")
> No help files found with alias or concept or title matching 'convex
> programming' using fuzzy matching.
> ...
>
>
> > Thanks,
> > Roger
> >
> > On 10/3/05, Prof Brian Ripley <ripley@stats.ox.ac.uk> wrote:
> >>
> >> On Mon, 3 Oct 2005, Huntsinger, Reid wrote:
> >>
> >>> Have you looked at the R interface to GLPK (the GNU Linear Programming
> >> Kit)?
> >>> http://cran.r-project.org/src/contrib/Descriptions/glpk.html
> >>
> >> NUOPT is not just about LP: the subject was `language for
> optimization'.
> >> Its manual says
> >>
> >> `NUOPT is a collection of powerful optimization methods, including:
> >>
> >> - primal-dual interior point method with higher order correction for
> >> Linear Programming (LP) models.
> >>
> >> - simplex method for Linear Programming (LP) and mixed integer
> programming
> >> (MILP) models.
> >>
> >> - primal-dual interior point method based on line search for general
> >> Convex Programming (CP) models including convex Quadratic Programming
> >> (CQP) models.
> >>
> >> - primal-dual interior point method based on trust region method for
> >> general Non-Linear Programming (NLP) models.
> >>
> >> - primal-dual interior point method based on quasi-Newton method for
> >> general Non-Linear Programming (NLP) models.
> >>
> >> - active set method for convex Quadratic Programming (CQP) models and
> >> mixed integer Quadratic Programming(MIQP) models.'
> >>
> >> In any case, I don't see GLPK as a `language' and other LP solvers are
> >> available in R.
> >>
> >>
> >>> From: r-help-bounces@stat.math.ethz.ch
> >>> [mailto:r-help-bounces@stat.math.ethz.ch] On Behalf Of Prof Brian
> Ripley
> >>> Sent: Sunday, October 02, 2005 10:38 AM
> >>> To: Paolo Cavatore
> >>> Cc: r-help@stat.math.ethz.ch
> >>> Subject: Re: [R] modeling language for optimization problems
> >>>
> >>>
> >>> On Sun, 2 Oct 2005, Paolo Cavatore wrote:
> >>>
> >>>> Does anyone know whether R has its own modeling language for
> >> optimization
> >>>> problems (like SIMPLE in NuOPT for S-plus)?
> >>>
> >>> No. Note that SIMPLE is the language of NUOPT, not of S-PLUS. There is
> >>> an (extra-cost) interface module S+NUOPT, but it is an interface to
> >>> NUOPT's engine.
> >>>
> >>> As far as I am aware R itself covers almost none of the ground of
> >> S+NUOPT,
> >>> and available packages cover only a small part of it.
> >>
> >> --
> >> 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 272866 (PA)
> >> Oxford OX1 3TG, UK Fax: +44 1865 272595
> >>
> >> ______________________________________________
> >> R-help@stat.math.ethz.ch mailing list
> >> https://stat.ethz.ch/mailman/listinfo/r-help
> >> PLEASE do read the posting guide!
> >> http://www.R-project.org/posting-guide.html
> >>
> >
> > [[alternative HTML version deleted]]
> >
> > ______________________________________________
> > R-help@stat.math.ethz.ch mailing list
> > https://stat.ethz.ch/mailman/listinfo/r-help
> > PLEASE do read the posting guide!
> http://www.R-project.org/posting-guide.html
> >
>
> --
> 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 272866 (PA)
> Oxford OX1 3TG, UK Fax: +44 1865 272595
>

        [[alternative HTML version deleted]]



R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html Received on Thu Oct 06 22:37:21 2005

This archive was generated by hypermail 2.1.8 : Sun 23 Oct 2005 - 18:24:46 EST