Re: R-beta: Help: cov.mve in R? dgamma in Splus?

Kurt Hornik (Kurt.Hornik@ci.tuwien.ac.at)
Mon, 23 Feb 1998 09:08:57 +0100 (CET)


From: Kurt Hornik <Kurt.Hornik@ci.tuwien.ac.at>
Date: Mon, 23 Feb 1998 09:08:57 +0100 (CET)
To: Peter Baker <p.baker@fsc.qut.edu.au>
Subject: Re: R-beta: Help: cov.mve in R? dgamma in Splus?
In-Reply-To: <Pine.OSF.3.95.980223141138.32070B-100000@einstein.math.sci.qut.edu.au>

>>>>> Peter Baker writes:

> Hi all
> I have a couple of obscure questions for R/Splus experts (which
> unfortunately isn't me!)

> I am trying to compute Bayes Factors using some Splus code of Raftery
> in Gilks et al (1996). Only problem is

> 1) R doesn't seem to have a robust covariance (cov.mve) which I
> suspect I need rather than a non-robust classical estimate

> 2) Splus has cov.mve BUT dgamma in Splus doesn't have a scale
> parameter (like in R). I presume it is easy to include one but it is
> not immediately obvious to me.

> I would prefer to use R and could probably modify minvol.f to compute
> cov.mve from R but don't want to reinvent the wheel.

It would definitely be great if you could provide code for cov.mve().

-k
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help 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-help-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._