Kriging Model Types

4677
2
08-21-2012 09:13 AM
DeanMorgan
New Contributor
Hi,

I have found that the model type used in Kriging can have a large influence on the RMSE of my model, but don't really understand how they work. For example, having ran lots of cross-validations with different data sets, I've found the K-Bessel model to yield the lowest RMSE, but cannot really explain why. I would like to briefly comment on the fundamentals of these models but cannot seem to find much information. Any links/references would be a great help, thank!

Dean
0 Kudos
2 Replies
EricKrause
Esri Regular Contributor
http://dusk.geo.orst.edu/gis/geostat_analyst.pdf

The formulas for the semivariogram models are in Appendix A, page 256 (page 262 of the pdf).  Each model corresponds to a different underlying physical process.

K-Bessel and Stable often produce the best results because they take an additional shape parameter that allows the models to change curvature while still maintaining the same nugget, range, and sill.
0 Kudos
DeanMorgan
New Contributor
Eric, prolific answers as usual, thank you!
0 Kudos