On Cokriging, Neural Networks, and Spatial Blind Source Separation for Multivariate Spatial Prediction

Multivariate measurements taken at irregularly sampled locations are a common form of data, for example, in geochemical analysis soil. In practical considerations, predictions these unobserved great interest. For standard multivariate spatial prediction methods it is mandatory to not only model dependencies but also cross-dependencies which makes demanding task. Recently, blind source separation (BSS) approach data was suggested. When using this BSS (SBSS) method before the actual prediction, modeling avoided, turn simplifies task significantly. letter, we investigate use SBSS as preprocessing tool and compare with from Cokriging neural networks an extensive simulation study well set.