[R] Oja median

[Rahul-A.Agarwal at ubs.com]

Apologies  Roger, if you find anything wrong with the mail.
Thanks for the reference Roger.
IS there anything else which I can look at, say something in R itself?
Regards and thanks once again

Rahul Agarwal 
Analyst 
Equities Quantitative Research 
UBS_ISC, Hyderabad 
On Net: 19 533 6363 




-----Original Message-----
From: roger koenker [mailto:rkoenker at uiuc.edu] 
Sent: Wednesday, November 19, 2008 10:49 PM
To: Agarwal, Rahul-A
Cc: r-help; R-SIG-robust at r-project.org
Subject: Re: [R] Oja median

Cross posting is sociopathic.  The Oja median _is_ robust in several
reasonable senses, but admittedly not from a breakdown perspective.
There are several other options:  for a good review see, e.g.
Multivariate Median, A. Niinimaa and H. Oja Encyclopedia of Statistical
Sciences


url:    www.econ.uiuc.edu/~roger            Roger Koenker
email    rkoenker at uiuc.edu            Department of Economics
vox:     217-333-4558                University of Illinois
fax:       217-244-6678                Champaign, IL 61820



On Nov 19, 2008, at 5:40 AM, <Rahul-A.Agarwal at ubs.com> wrote:

>
> Hi Roger,
>
> As we know that The Oja median has (finite) breakdown point 2/n, i.e.,

> is not robust in any reasonable sense, and is quite expensive to 
> compute, so do we have some better methodology to compute multivariate

> median
>
>
> Rahul Agarwal
> Analyst
> Equities Quantitative Research
> UBS_ISC, Hyderabad
> On Net: 19 533 6363
>
>
>
>
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> 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.