[R-sig-Geo] YNT: Complete spatial randomness testing

[Cenk İÇÖZ]

According to your suggestion , I excluded duplicated points. However I took the same error message.

udp1<-unique.ppp(dp1)
 xxx<-allstats(udp1)
Error: in Fest(X, r) the successive r values must be finely spaced: given spacing = 0.010196; required spacing <=  0.00586

I attached the data for you to investigate. Thanks a lot. 

Cenk 

-----Original Message-----
From: Rolf Turner [mailto:r.turner at auckland.ac.nz] 
Sent: Sunday, October 2, 2016 10:17 PM
To: Cenk İÇÖZ <cicoz at anadolu.edu.tr>
Cc: Adrian.Baddeley at curtin.edu.au; Ege Rubak <rubak at math.aau.dk>
Subject: Re: YNT: [R-sig-Geo] Complete spatial randomness testing

On 02/10/16 23:42, Cenk İÇÖZ wrote:
> Hi Rolf,
>
> Thanks for the reply.
>
> A) I updated my R version and spatstat package. Currently  using  spatstat 1.46-1 spoiler alert and  3.3.1 version of R.
>
> B) Yes the distributions of interpoint distancess. Sorry to mention it wrong. In some sources they call distance based tests on complete spatial randomness.
>
> C)  Again yes. dp1  is a class of ppp object of spatstat constructed like this.
> Latitude and longtitude and the specific window of the study area.
>
> dp1<-ppp(veri$Enlem, veri$Boylam, c(30,40), c(39,42))
>
> D) I also tried constructing the plot with a simulated pattern. No problem in plotting the graph.
>
> E) I performed the function Jest and plotted it . I took the same error message again unfortunately.
>
> """Error: in Fest(X, r) the successive r values must be finely spaced: given spacing = 0.010196; required spacing <=  0.00586""""
>
> My question is that I have duplicated points of earthquake occurrences in the study area. Is it a problem related to  it?
>  Or the spacing is too low to draw the plot that function can not manage to draw  it such a short distance.
>
> Thanks again all for the all help .
>

The duplicated points could indeed be the problem.  Try

udp1 <- unique(dp1)
xxx  <- allstats(udp1)
plot(xxx)

If that doesn't work, can you make the data set "veri" available to us so that we can investigate?

cheers,

Rolf

--
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276
-------------- next part --------------
A non-text attachment was scrubbed...
Name: veri.RData
Type: application/octet-stream
Size: 8661 bytes
Desc: veri.RData
URL: <https://stat.ethz.ch/pipermail/r-sig-geo/attachments/20161003/7588d9e5/attachment.obj>