sampling with a "cluster" constraints

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

sampling with a "cluster" constraints

Patrick Giraudoux
Dear Listers,

Everybody is probably used with sp:spsample, which permits to produce
points locations at random or on a regular grid. Is anybody aware of a
package permitting similarly to produce points locations in a field with
various degrees of clustering. The idea is to produce points in a field
with various degrees of clustering repetedly,  and to compare statistics
to observed data.  E.g. observations in the real world are clustered;
one computes a statistics, e.g. the distance of each observation to a
feature (e.g. to a road network); then one compute B=999 similarly
clustered distributions randomly and the corresponding statistics. This
would allow to compute how likely the observed statistics is.

Any hint welcome!

Best,

Patrick


        [[alternative HTML version deleted]]

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: sampling with a "cluster" constraints

Roger Bivand
Administrator
Just a thought - spatstat? This has had line segements too for some time,
maybe chapter 17 in the spatstat book?

Roger

On Thu, 22 Mar 2018, Patrick Giraudoux wrote:

> Dear Listers,
>
> Everybody is probably used with sp:spsample, which permits to produce
> points locations at random or on a regular grid. Is anybody aware of a
> package permitting similarly to produce points locations in a field with
> various degrees of clustering. The idea is to produce points in a field
> with various degrees of clustering repetedly,  and to compare statistics
> to observed data.  E.g. observations in the real world are clustered;
> one computes a statistics, e.g. the distance of each observation to a
> feature (e.g. to a road network); then one compute B=999 similarly
> clustered distributions randomly and the corresponding statistics. This
> would allow to compute how likely the observed statistics is.
>
> Any hint welcome!
>
> Best,
>
> Patrick
>
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-Geo mailing list
> [hidden email]
> https://stat.ethz.ch/mailman/listinfo/r-sig-geo
>
--
Roger Bivand
Department of Economics, Norwegian School of Economics,
Helleveien 30, N-5045 Bergen, Norway.
voice: +47 55 95 93 55; e-mail: [hidden email]
http://orcid.org/0000-0003-2392-6140
https://scholar.google.no/citations?user=AWeghB0AAAAJ&hl=en
_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Roger Bivand
Department of Economics
Norwegian School of Economics
Helleveien 30
N-5045 Bergen, Norway
Reply | Threaded
Open this post in threaded view
|

Re: sampling with a "cluster" constraints

Patrick Giraudoux
Le 23/03/2018 à 08:59, Roger Bivand a écrit :
> Just a thought - spatstat? This has had line segements too for some
> time, maybe chapter 17 in the spatstat book?
>
> Roger

Thanks Roger. I will check this. If nothing comes out, I am thinking
about developping a quick and dirty home-made function based on negative
binomial distribution where a "clustering" (dispersion) parameter can be
tuned... Hence, shit stats for hazel grouse and capercaillie shit
distribution :-D...

Patric

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo
Reply | Threaded
Open this post in threaded view
|

Re: sampling with a "cluster" constraints

Patrick Giraudoux
In reply to this post by Roger Bivand
Le 23/03/2018 à 08:59, Roger Bivand a écrit :
> Just a thought - spatstat? This has had line segements too for some
> time, maybe chapter 17 in the spatstat book?
>
> Roger

Indeed. There are at least two possibilities: simulations using either
Matern process or Neyman-Scott Point Process with Variance Gamma cluster
kernel... Will see how it could be applicable in my case. Thanks again
for the hint, Roger.

Patrick

_______________________________________________
R-sig-Geo mailing list
[hidden email]
https://stat.ethz.ch/mailman/listinfo/r-sig-geo