On 15.05.2014 15:47, Edzer Pebesma wrote: > On 05/15/2014 10:58 AM, Roelof Coster wrote: >> Also, in case kriging in a neighbourhood is possible, how does the function >> determine this neighbourhood? How is distance in space compared to distance >> in time? > > It uses http://cran.r-project.org/web/packages/FNN/index.html > to get nearest neighbours, and, unless parameter stAnis is specified, > tries to figure out a sensible space/time anisotropy from the variogram > model. The source is found > https://r-forge.r-project.org/scm/viewvc.php/pkg/R/krige0.R?view=markup&root=gstat > and look for function krigeST.local
The "closest" neighbours are the "strongest" correlated ones. To achieve this, a larger (4-fold by default) metric spatio-temporal neighbourhood with some spatio-temporal anisotropy (stAni or from the variogram model) is selected. For these neighbours, the spatio-temporal covariance function is evaluated and the ones with the strongest correlation are selected. The interpolation is then carried out using these nmax neighbours. HTH, Ben > > The curent version on CRAN does not pass through parameters > checkNeighbourhood and bufferNmax, just corrected this on r-forge. > > Wbr, > >> >> Thanks! Roelof Coster >> >> [[alternative HTML version deleted]] >> >> _______________________________________________ >> R-sig-Geo mailing list >> R-sig-Geo@r-project.org >> https://stat.ethz.ch/mailman/listinfo/r-sig-geo >> > > > > _______________________________________________ > R-sig-Geo mailing list > R-sig-Geo@r-project.org > https://stat.ethz.ch/mailman/listinfo/r-sig-geo > -- Benedikt Gräler ifgi - Institute for Geoinformatics University of Muenster http://ifgi.uni-muenster.de/graeler Phone: +49 251 83-33082 Mail: ben.grae...@uni-muenster.de
signature.asc
Description: OpenPGP digital signature
_______________________________________________ R-sig-Geo mailing list R-sig-Geo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo