[R] Resampling Help Needed

KoopaTrooper ncooper1 at tulane.edu
Tue Dec 4 00:53:54 CET 2012


I am using package ks() to build 3D representations of bird territories and
calculate territory volume from spatial data (simply x, y, and z
coordinates). What I want to do is determine at what sample size (#
locations collected) does the territory volume stop increasing. This should
give me an idea of the number of points needed for future seasons. 

So I have a couple of birds each with 200 spatial locations (x,y,z). I want
to run the following code (see below), but have R calculate territory size
100 times with 10 random points (no replacement), 100 times with 20 random
points, 100 times with 30 random points, etc. I can figure out how to do
this manually (i.e. create 100 individual files with 10 random points, 20
random points, etc.) but I figure there must be a way to make my life
easier. Any help would be appreciated. Even pointing me in the correct
direction would be a big help. Thanks!

Nathan

#read data files (.csv's with 200 rows of x,y,z coordinates)
a<-read.csv("A_PW_ASY_M_LII_2011.csv")

#calls the plug-in bandwidth estimator
Ha <- Hpi(a) 

#sets min/max grid size for each dimension
minX<-min(a$X)-25
minY<-min(a$Y)-25
minZ<-0

maxX<-max(a$X)+25
maxY<-max(a$Y)+25
maxZ<-max(a$Z)+5

#creates kernel utilization distribution
fhata <- kde(x=a, H=Ha, binned=FALSE, xmin=c(minX,minY,minZ),
xmax=c(maxX,maxY,maxZ)) 

#calculates territory volume at 95% isopleth
Vol95<-contourSizes(fhata, cont=95)



--
View this message in context: http://r.789695.n4.nabble.com/Resampling-Help-Needed-tp4651973.html
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list