R help - Sep 2008 - Oja median

Hi,
 
Can we get the code for calculating Oja median for multivariate data
 
Thanks and Regards

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



 

	[[alternative HTML version deleted]]

> Can we get the code for calculating Oja median for multivariate data
RSiteSearch("oja median") returns a link to this R-help post with code
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/12781.html

Regards,
Richie.

Mathematical Sciences Unit
HSL


------------------------------------------------------------------------
ATTENTION:

This message contains privileged and confidential inform...{{dropped:20}}

Hi Richard,

Thanks for the code.....but I already have a code for bivariate data...I
want it for data when it is multivariate or its dimension is greater
than two...i mean to say if it is trivariate or higher dimension....can
I get a more generalised form of oja median code

Thanks and Regards  


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




-----Original Message-----
From: Richard.Cotton at hsl.gov.uk [mailto:Richard.Cotton at hsl.gov.uk] 
Sent: Thursday, September 18, 2008 3:02 PM
To: Agarwal, Rahul-A
Cc: r-help at r-project.org; r-help-bounces at r-project.org
Subject: Re: [R] Oja median
> Can we get the code for calculating Oja median for multivariate data
RSiteSearch("oja median") returns a link to this R-help post with code
http://finzi.psych.upenn.edu/R/Rhelp02a/archive/12781.html

Regards,
Richie.

Mathematical Sciences Unit
HSL


------------------------------------------------------------------------
ATTENTION:

This message contains privileged and confidential inform...{{dropped:20}}

>>>>>   <Rahul-A.Agarwal at ubs.com>
>>>>>     on Thu, 18 Sep 2008 05:20:56 -0400 writes:
    > Hi,
 
    > Can we get the code for calculating Oja median for
    > multivariate data

Excuse me, but must you really?

The Oja median has (finite) breakdown point 2/n, i.e., is not
robust in any reasonable sense, and is quite expensive to compute, etc etc.

Rather use a better robust method,
typically (start with)  cov.rob() from the recommended (hence
part of every correct R installation) MASS package.

There are (somewhat) better methods available for robust
"location+scatter" [aka "(mu , Sigma)"] from packages 
'rrcov', 'robustbase' and/or 'robust', e.g.,
robustbase::covMcd

If you need to know more, please divert to the 
SIG (special interest group) mailling list  R-SIG-robust
to which I CC this.

Martin Maechler, ETH Zurich