[R] MCD distance

[Prof Brian Ripley]

There first two *are* in the lqs package: you just need to use 
mahalanobis() with cob.rob().

I've never heard of BACON.

On Fri, 6 Jun 2003, osama wrote:

>     I would like to compare robust distance measures,e.g. MVE, MCD,
> BACON, as measures of leverage and to detect outliers in X-space . I
> could not find it in LQS or MASS backages.

Why look in only two packages?

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595