[S] multivariate outliers in regression

John P McKinlay (john.mckinlay1@jcu.edu.au)
Sat, 14 Feb 1998 09:50:11 +1000


Can anyone share an S+, C or fortran implementation of the robust distance
methodology of Rousseeuw and van Zomeren (1990)? I would be interested in
hearing from anyone who uses this methodology, and in particular I'd like
to hear opinions about the critical comments of Cook and Hawkins. A casual
search of the literature has not turned up anything on this method since
the original 1990 paper - any pointers?

Ref: P Rousseeuw and B van Zomeren (1990) Unmasking Multivariate Outliers
and Leverage Points (with comment)
JASA 85(411):633-51

Thanks...

------------------------------------------------------------
John McKinlay

CRC Reef Research Centre
James Cook University of North Queensland
Townsville 4811
Australia
Phone: +61 07 47716146 or +61 07 47815253
Fax: +61 07 47814099
Email: john.mckinlay1@jcu.edu.au
WWW: http://www.gbrmpa.gov.au/~crcreef/
------------------------------------------------------------
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news