[R-sig-Geo] Ordinary and Regression Kriging combined to deal with missing values in predictor variables

[piero campa]

Dear list,
a slightly different scenario: what if one has a set of ubiquitous
predictors and e.g. one predictor which has a limited number of
observations?

Hengl in his book "A practical guide to geostatistical mapping" suggests
(probably) to combine RK with CK so that "additional, more densely sampled
covariates, can be used to improve spatial interpolation of the residuals.
The interpolated residuals can then be added to the deterministic part of
variation".

I cannot figure out what this do imply in detail sincerely: I have my
regression model with the ubiquitous predictors (and hence the corresponding
residuals); I have the residuals of the CK with the non-ubiquitous
predictor. How to merge the information?

Thank you in advance,
Piero
-- 
View this message in context: http://r-sig-geo.2731867.n2.nabble.com/Ordinary-and-Regression-Kriging-combined-to-deal-with-missing-values-in-predictor-variables-tp2764942p6022991.html
Sent from the R-sig-geo mailing list archive at Nabble.com.