CMStatistics 2017: Start Registration
View Submission - CMStatistics
B0552
Title: Isotropic variogram models on all spheres Authors:  Juan Du - Kansas State University (United States) [presenting]
Abstract: Variogram or variogram matrix functions play an important role in modeling dependence structure among multiple processes at different locations in spatial statistics. The aim is to characterize the variogram models on spheres of all dimensions for intrinsically stationary, isotropic, univariate, and multivariate processes. Efficient approaches are proposed to construct a variety of isotropic variogram functions including simple polynomial structures. Series representations and spherical behavior of intrinsic stationary random fields are elaborated in both theoretical and simulation study. Applications of the proposed model and related theoretical results are demonstrated using simulation and real data analysis.