[R] About the efficiency of R optimization function

cls59 sharpsteen at mac.com
Sat May 16 01:12:20 CEST 2009

popo UBC wrote:
> Hi Charlie,
> Thank you so much for suggestions!!
> Actually, I used the optimization toolbox in MABLAB before and I even
> wrote
> some numerical optimization programs by myself. As far as I know, some
> commercial optimization softwares had already replaced L-BFGS-B by more
> advanced algorithms, such as interior point method, SQP(sequential
> quadratic
> programming), implemented under trust region strategy. So,
>    - Have you ever tried these techniques? Are they available in R
> already?
>    - In your previous experieces, did R work satisfactory? I mean, was it
>    often that R failed to converge or spent too much time?
>    - Mainly, I need to calculate the MLE. But I really have no idea what
> the
>    likelihood may looks like. According to your experiences, would the
>    likelihood function be too complicated? Is L-BFGS-B good enough?
>  Thanks again!!
> Popo

Hi Popo,

Truthfully, I do not have much direct experience with the R optim function.
For my linear optimization needs I usually use a package called HOPDM which
implements an interior point method. For nonlinear problems I use MINOS
which relies on conjugate gradients and interfaces nicely with the
simulation models I use in my field of study.

I have however taken a groundwater course where the class was assigned an
optimization problem concerning water supply. The objective function was
nonlinear and contained tens or hundreds of variables depending on the level
of discretization used in the numerical model. Some students used MINOS,
others used MATLAB and a few used R. All three packages generated similar

If you are solving a maximum likelyhood problem, check out the mle function
in the stats4 package- it appears to be a front-end to optim which is
designed to tackle these sorts of problems. And, as I noted before, it is
possible to call a C or Fortran optimizer that you trust from within R.
Compared to my experience with MATLAB, calling an external Fortran routine
was an absolute walk in the park with R.

Good luck!


Charlie Sharpsteen
Environmental Resources Engineering
Humboldt State University
View this message in context: http://www.nabble.com/About-the-efficiency-of-R-optimization-function-tp23552061p23568607.html
Sent from the R help mailing list archive at Nabble.com.

More information about the R-help mailing list