R. Giraldo Henao, J. Mateu Mahiques, A. Reyes Lizarazo

Recently several methodologies for carrying out geostatistical analysis of functional data has been proposed.All of them assume that the spatial functional process considered is stationary. However in practice often we have non-stationary functional data sets because there is spatial trend in the mean. Here we propose a methodology to extend kriging predictors for functional data to the case where the mean function is not constant through the region. We consider an approach based on the classical residual kriging method used in univariate geostatistcs. We propose a three steps procedure. Initially a functional regression model is used for detrending the mean. Posteriorly we apply some kriging method for functional data to the regression residuals for doing prediction of a residual curve on a non-data location. Finally the prediction curve is obtained as the sum of the trend and the residual prediction.

Palabras clave: functional data kriging stationarity

F. J. Esquivel Sánchez, J. M. Angulo Ibáñez

P. Juan Verdoy, C. Díaz-Ávalos, J. Mateu Mahiques