[R-sig-Geo] Complete spatial randomness testing

[Rolf Turner]

On 02/10/16 02:02, Cenk İÇÖZ via R-sig-Geo wrote:
> Hi friends,
>
> I have a point pattern that consist of earthquake locations.  I want
> to test the pattern for complete spatial randomness based on distance
> functions. I was using spatstat package allstats function to plot 4
> of them together in a graph  however I took this error message:
>
>
> "  Error in plot(allstats(dp1)) : error in evaluating the argument
> 'x' in selecting a method for function 'plot': Error: in Fest(X, r)
> the successive r values must be finely spaced: given spacing =
> 0.010196; required spacing <=  0.00586   "
>
>
> Is it something related to plotting window properties? Although I got
> an error about F function , I could plot it individually. In addition
> I could plot  all G, F, K  and L functions together in a graph
> manually. Also I could not find a function to estimate J function in
> spatstat package. It is only included in allstat function.
>
> Thanks in advance .

(a) What version of spatstat are you using?  It may be out of date.  The 
version of spatstat currently on CRAN is 1.46-1.

(b) When you say "based on distance functions" I presume that you mean 
"based on various distributions of interpoint distances".

(c) I presume that "dp1" is the point pattern (object of class "ppp") of 
earthquake locations.

(d) You have not provided a reproducible example of your problem. 
Consequently I checked things out with a simulated example:

set.seed(42)
X <- rpoispp(100)
plot(allstats(X))

This ran with no problem, and produced the expected plot.

(e) The function to estimate the J function is Jest().  It has been in 
the spatstat package for a very long time.

cheers,

Rolf Turner

-- 
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276