A new procedure is developed for spatial sampling designs. Based on literature on similar studies, the optimal sampling design for spatial prediction with estimated parameters is nearly regular with a few clustered points. Since the pattern is similar to a generalization of the Neyman-Scott (GNS) process which allows for regularity in the parent process, this article proposes use of the realization of the GNS process as sampling design points. This method translates the high-dimensional optimization problem of selecting sampling sites into a low-dimensional optimization problem of searching for optimal parameter sets in the GNS process. The sampling design algorithm that uses the GNS process is computationally more efficient than traditional methods while achieving similar minimization of the criterion functions. While the traditional methods become computationally infeasible for a sample size larger than a hundred, the proposed algorithm is applicable to a larger sample size. A real dataset is used with the proposed algorithm to find the optimal spatial design for predicting sea surface temperature in the Pacific Ocean.
展开▼
