Spatial data, rpoispp, using window with fixed radius?

What's wrong with rpoispp in spatstat? It can simulate over a polygon,
which can of course be used to closely approximate a circle. There is also
spsample in the sp package.

I'd also suggest asking this question on r-sig-geo.

Don MacQueen

Lawrence Livermore National Laboratory
7000 East Ave., L-627
Livermore, CA 94550
925-423-1062





On 1/6/12 5:17 AM, "herbert8686 at gmx.de" <herbert8686 at gmx.de> wrote:

>Dear All,
>
>I was searching through the spatstat manual in order to find a function
>to simulate a Poisson pattern only within a fixed radius (circular moving
>window) around individual points. If points are distributed
>heterogeneously over a large area this may help to only assess deviation
>from CSR within the window and thus does not require additional
>information on a covariate. I could not find such a function in spatstat.
>Can please anyone help?
>
>Thanks,
>
>Herb
>
>--
>
>______________________________________________
>R-help at r-project.org mailing list
>https://stat.ethz.ch/mailman/listinfo/r-help
>PLEASE do read the posting guide
>http://www.R-project.org/posting-guide.html
>and provide commented, minimal, self-contained, reproducible code.

Dear All,

I was searching through the spatstat manual in order to find a function to simulate a Poisson pattern only within a fixed radius (circular moving window) around individual points. If points are distributed heterogeneously over a large area this may help to only assess deviation from CSR within the window and thus does not require additional information on a covariate. I could not find such a function in spatstat. Can please anyone help?

It's not clear to me just what you want to do, but it *sounds* like you want
to simulate a cluster process with each cluster being a Poisson pattern
in a disk of fixed radius.  If so, the function rMatClust() does just 
what you
want.

It seems to me also possible that you want to "pre-specify" the cluster 
centres,
or parent points.  Such a capability is *not* currently built into 
"spatstat" but
would not be hard to code up.  Let me know if you indeed want to pre-specify
the cluster centres.

     cheers,

         Rolf Turner

The following message appeared on R-help but this discussion should be moved to R-sig-geo

I was searching through the spatstat manual in order to find a function
 to simulate a Poisson pattern only within a fixed radius (circular moving window)

> around individual points. If points are distributed heterogeneously over a

large area this may help to only assess deviation from CSR within the window 
and thus does not require additional information on a covariate. 
I could not find such a function in spatstat. Can please anyone help?

What do you want to happen if two of the circles overlap? Should the density of random points be twice as high? 

If the answer is 'yes' then do the following (where X is your original point pattern of centres, and 'r' is the radius of the circles, and 'lambda' is the intensity of random points per unit area in each circle)

       V <- scanmeasure(X, r)
       V <- eval.im(lambda * V)
       Y <- rpoispp(V)

If the answer is 'no' then do 
       W <- dilation(X, r)
       Y <- rpoispp(lambda, win=W)

Adrian Baddeley

An embedded and charset-unspecified text was scrubbed...
Name: not available
URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20120110/55cff976/attachment.pl>