On 02/13/2014 01:30 PM, Roelof Coster wrote: > Hi all, > > I want to do a spatio-temporal regression on a quite large dataset. I have >> 100 k records. These correspond to measurements taken at 3000 locations, > approximately every half year. The geographic area is all of the > Netherlands (240 x 300 km).
Sample variograms are a quadratic form, meaning that all pairs of your 100K records are evaluated; that makes 10^10 pairs. Are you a patient person? > Is spatio-temporal kriging advisable for a dataset that is so large? > > When I make the sample space-time variogram (with variogramST), it > automatically chooses a time-lag difference of about 2 days. This is much > too small to be meaningful for my data; half-year periods would be > interesting. Is there a way to tell this to the variogramST function? variogramST (in package gstat) assumes that the time gap between observations is constant. If you have "approximately every half year" observations, are some of them only a few days apart? Did you see a warning printed? If the time differences are very un-even, the whole activity may be pretty meaningless (with current implementation), and a mapping to stricter regularity might help. We have coding Friday tomorrow, drop by or send something useful and we might be able to work it out. -- Edzer Pebesma Institute for Geoinformatics (ifgi), University of Münster Heisenbergstraße 2, 48149 Münster, Germany. Phone: +49 251 83 33081 http://ifgi.uni-muenster.de GPG key ID 0xAC227795
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