[R] help on moran's I index of point pattern, not areal pattern
Roger.Bivand at nhh.no
Tue Mar 14 20:38:44 CET 2006
On Wed, 15 Mar 2006, zhijie zhang wrote:
> we all know that moran's I index and Geary'C index can be used to test
> spatial autocorrelation in both the area data and point data, but i only can
> find something on how to calculate on the data of area, and can't find the
> methods to perform it on the point data, could anybody give me some
> information, thanks in advance!
There is no difference between area and point data given a
Voronoi/Dirichlet tesselation. To use any such index, you are obliged to
define neighours anyway, and whether you use a distance criterion,
k-nearest neighbours, triangulation, or a graph criterion such as a
minimum spanning tree, the duality remains.
Functions for defining neighbours are found in ade4 and spdep, and suit
points at least adequately. If you want a distance-based correlogram,
define increasing distance bands (or see the off-CRAN ncf package).
The functions are (among others) mstree() in ade4 taking a distance matrix
as its argument, and dnearneigh(), knearneigh(), tri2nb(), gabrielneigh(),
relativeneigh(), and soi.graph() taking a matrix of 2D coordinates as
their argument in spdep (if your points are a regular grid, there is also
cell2nb()). At least you have a choice here.
> Kind Regards,
> Zhi Jie,Zhang ,PHD
> Department of Epidemiology, School of Public Health, Fudan University
> [[alternative HTML version deleted]]
> R-help at stat.math.ethz.ch mailing list
> PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Economic Geography Section, Department of Economics, Norwegian School of
Economics and Business Administration, Helleveien 30, N-5045 Bergen,
Norway. voice: +47 55 95 93 55; fax +47 55 95 95 43
e-mail: Roger.Bivand at nhh.no
More information about the R-help