[6359] in s-news-athena
Re: [Q] is there a 2d analogue of density()?
daemon@ATHENA.MIT.EDU (Martin Maechler)
Thu Feb 2 04:31:42 1995
From: Martin Maechler <maechler@stat.math.ethz.ch>
Date: Thu, 2 Feb 1995 10:11:32 +0100
To: ross@psy.uq.oz.au
Cc: s-news@utstat.toronto.edu
In-Reply-To: <3gosc2$27n@dingo.cc.uq.oz.au> (ross@psy.uq.oz.au)
Reply-To: Martin Maechler <maechler@stat.math.ethz.ch>
Cc: maechler@stat.math.ethz.ch
>>>>> "Ross" == Ross Gayler <ross@psy.uq.oz.au> writes:
Ross> I have a data set with 2 predictor variables and would like to
Ross> know the distribution of observations over the plane defined by
Ross> the predictors. I can easily divide the plane into bins and
Ross> count the observations but would prefer smoothed estimates (as
Ross> provide by density() in the 1d case).
Ross> Could anyone point me at an off-the-shelf function to do this?
Ross> I am using S-Plus for Windows V3.2
Yes, our 'kde2d' submission on statlib in the S directory.
It does 2D Gaussian kernel density estimation
[INCLUDING 'spherizing' which may be important for
strongly correlated data]
it is 'plain S'...
Martin Maechler <maechler@stat.math.ethz.ch> <>< _
Seminar fuer Statistik, SOL F5 _| |_
ETH (Federal Inst. Technology) 8092 Zurich SWITZERLAND |_ _|
phone: x-41-1-632-3408 fax: ...-1086 |_|
