I want to perform cell declustering in a small point data set.
I have 9 observation points and I want to find the size of the region of influence for an observation, and afterwards use that as weight in analysis.
The only way, I could find is that one:
library('gstat') library('sp') data(meuse) coordinates(meuse) = ~x + y data(meuse.grid) coordinates(meuse.grid) = ~x + y gridded(meuse.grid) <- TRUE zn.tp = krige(log(zinc) ~ 1, meuse, meuse.grid, nmax = 1) image(zn.tp["var1.pred"]) points(meuse, pch = "+", cex = 0.65) cc = coordinates(meuse) library(tripack) plot(voronoi.mosaic(cc[, 1], cc[, 2]), do.points = FALSE, add = TRUE) title("Thiessen (or Voronoi) polygon interpolation of log(zinc)")
So, after preparing the data, I could create Thiessen (Voronoi) polygons. However, It works properly only when I made decision about the borders of my grid.
How I could perform cell declustering, that will also take care of creating the outer borders?
The only idea I could find was DECLUST function from gslib http://www.statios.com/help/declus.html . Could cell declustering be done in less difficult way?