Point-wise Kriging for Spatial Prediction of Functional Data
Pedro Delicado, Ramón Giraldo, Jorge Mateu · Contributions to statistics · 2008
We propose a methodology to carry out spatial prediction when measured data are curves. Our approach is based on both the kriging predictor and the functional linear point-wise model theory. The spatial prediction of an unobserved curve is obtained as a linear combination of observed functions. We employ a solution based on basis function to estimate the functional parameters. A real data set is used to illustrate the proposals.