[R-sig-Geo] Characterizing the position (central vs > peripheral) of a point in a point pattern

[Nicholas Lewin-Koh]

Hi Alex
I think the following might be close to what you want;
1) calculate the convex hull of your point set
2) Calculate the centroid of the convex hull polygon
3) for each point:
   calculate the distance to the centroid (d_c)
   calculate the distance to the closest boundary (d_b)
4) interior = d_c < d_b

If this simple 50/50 rule is to simple, you could use convex hull
peeling
to determine the exterior points for some threshold, eg the outer 30% of
the data.

hope this helps
Nicholas
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Wed, 29 Apr 2009 12:12:42 +0200
> From: Alexandre VILLERS <alexandre.villers at cebc.cnrs.fr>
> Subject: [R-sig-Geo] Characterizing the position (central vs
> 	peripheral) of a point in a point pattern
> To: Aide R SIG GEO <r-sig-geo at stat.math.ethz.ch>
> Message-ID: <49F8281A.9020608 at cebc.cnrs.fr>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
> 
> Good morning,
> 
> I would like to characterize the position of  a given nest in a 
> population of breeding birds. Does anyone know a statisitics (and 
> associated R package) that could objectively tell whether a nest has a 
> central or peripheral location ? It also needs to be density independent 
> (not affected by sample size)
> Hope this is a trivial question !
> 
> Thanks for your help.
> 
> Alex
> 
> -- 
> Alexandre Villers
> PhD Student
> Team "Biodiversity"
> Centre d'Etudes Biologiques de Chiz?-CNRS UPR1934
> 79360 Beauvoir sur Niort
> 
> Phone +33 (0)5 49 09 96 13
> Fax   +33 (0)5 49 09 65 26
> 
> 
> 
> 
> __________ Information from ESET Mail Security, version of virus
> signature database 4042 (20090429) __________
> 
> The message was checked by ESET Mail Security.
> http://www.eset.com
> 
>