Spatial semi-parametric bootstrap method for analysis of kriging predictor of random field

Abstract

In the spatial statistics it is often assumed that the data follow a Gaussian random field. Efron introduced bootstrap method for independent data analysis (IIDB) but it can not be applied in spatial data analysis because of dependency of observations. In this paper, an algorithm is given for spatial semi-parametric bootstrap (SSPB) method to estimate the precision measures of plug-in kriging predictor of random field. We also compare IIDB and SSPB methods for analysis of plug-in kriging predictor in a Monte-Carlo simulation study. Finally, we use SSPB method for analysis of finite strain data in geology. © 2010 Published by Elsevier Ltd. Ltd.