Dear Alex,
after installation of ROptRegTS there should be a folder "scripts" in
the package directory of ROptRegTS which includes some further examples.
If you are assuming normal errors you should switch to package RobRex
which is "optimized" for normal errors. There you will also find a
folder "scripts" in the package directory.
In general, multivariate regression should work. But, I have to admit
that we didn't try many multivariate examples so far. Beside that, the
user interface is still rudimentary at the moment and some ideas which
will clearly reduce computation time (as in the case of package RobLox)
are unfortunately not yet implemented.
So, if you are willing to accept these drawbacks I will be glad to give
you a hand to get our estimators to work for your data.
Best,
Matthias
On 06.06.2010 06:47, Alex Weslowski wrote:> Hi all,
>
> Other than:
> http://www.stamats.de/F2006.r
>
> Are there other good simple examples out there of using the ROptRegTS
> package (part of the RobASt project)? I'm hoping to plug it in for
> multivariate regression. Or is this not a good idea? Just trying to
> find out how it compares to rlm, lts, glm, etc. Hopefully this makes
> sense, I'm new to the world of statistics and R.
>
> Thanks!
>
> St0x
>
> ______________________________________________
> 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.
>
--
Dr. Matthias Kohl
www.stamats.de