[R-sig-Geo] Loosmore and Ford (2006) Goodness-of-fit (GoF) test for CSR

[Aurelie C.Godin]

Thank you Marcelino!

You are right, I am trying to test the random labelling hypothesis instead
of CSR.
I've been able to run your LF.gof with the nztrees dataset, however, when I
test the envelope with the simulated rlabel list and run the LF.gof I get an
error message:
http://r-sig-geo.2731867.n2.nabble.com/file/n7336880/OTGR.RData OTGR.RData 

> LF.gof(GR.env) 
Error in `[.data.frame`(as.data.frame(attributes(X)$simfuns), , 1) : 
  undefined columns selected

I've also tried the following coding for which I still get the same error
message.

# Precomputed list of point patterns
OTim<-as.im(OT)#create a pixel image of the point pattern of all my events
(OT)
# Simulate GR$n random point patterns within OTim
Xlist <- list()
for (i in 1:198) Xlist[[i]] <-  rpoint(563,OTim)
#compute the envelope
EE<-envelope(GR,Kest,simulate=Xlist)
> LF.gof(EE)
Error in `[.data.frame`(as.data.frame(attributes(X)$simfuns), , 1) : 
  undefined columns selected

I thought that the above code would give me a similar result then your code,
but I do get quiet different plots. Is this simply because of the pixel
image? Or is there something else here that I'm missing? 

Any ideas of what can cause this error message?
Thank you very much,


-----
Aurelie Cosandey-Godin
Ph.D. Student, Department of Biology, Dalhousie University
Industrial Graduate Fellow, WWF-Canada
Email: godina at dal.ca | Web: wormlab.biology.dal.ca

--
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Loosmore-and-Ford-2006-Goodness-of-fit-GoF-test-for-CSR-tp7332928p7336880.html
Sent from the R-sig-geo mailing list archive at Nabble.com.