Optimal prediction in isotropic spatial process under spherical type variogram model with application to corn plant data

• Published in: Journal of physics. Conference series 2021-06, Vol.1940 (1), p.12003
• Main Authors: Somayasa, W, Sutiari, D K, Sutisna, W
• Format: Article
• Language: English
• Subjects: Fertility; Geostatistics; Least squares method; Mathematical models; Physics; Spatial data; Weighting functions
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/1940/1/012003

A central problem in modelling of spatial data is the construction of an optimal prediction map of a numerical quantity or variable under study. In geostatistics this study is commonly called universal kriging. In this paper we study weighted universal kriging method in which a weight function is used for improving the efficiency of the predictor. We consider isotropic spatial process under a variogram function belongs to the parametric family of spherical type. The parameters of the postulated variogram model are estimated by applying ordinary least squares method in that the squared distance between the variogram model and the associated variogram sample is minimized. Numerical approximation for finding the solution of the least squares equations is conducted by using graphical approach. The application of the method to a corn plant data results in the kriging map of the maximum height that can be achieved by the corn plants planted over a rectangular farm land. This investigation result can in advance describe the map of the fertility level of the farm land where the corn have been planted.