Hi, thanks a lot for the explanation. My motivation is to do some sort of space-time bootstrapping. Probably this is a big misuse of the spacetime package as I loose the correct time but I want to use the procedure I am using for the original sample also for the bootstrapped ones. And at the moment I don't know another way for keeping the spatial dependencies. But the way with the logical vector is fine - and if you have any other suggestions (for the whole bootstrapping thing) please let me know.
Cheers, Martin _______________________________________________ R-sig-Geo mailing list R-sig-Geo@r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-geo