A0246
Title: TSSS: A novel triangulated spherical spline smoothing for data distributed on complex surfaces
Authors: Zhiling Gu - Iowa State University (United States) [presenting]
Shan Yu - University of Virginia (United States)
Guannan Wang - College of William & Mary (United States)
Lily Wang - George Mason University (United States)
Abstract: Data distributed on surfaces has been widely observed and analyzed in practice, especially in Earth, planetary, and biomedical science. Examples include the estimation of geopotential for the Earth, predicting the magnetic North Pole's movement, and completing the cosmic microwave background radiation field. A novel nonparametric method is introduced to efficiently discover the underlying signals on surfaces of complex domains. In particular, a penalized spline estimator defined on a triangulation of surface patches with irregular shapes is proposed, which guarantees signal matching and smoothness. Moreover, the asymptotic behaviour of the proposed estimators is investigated, which indicates the proposed estimation enjoys a desirable convergence. Simulation experiments and data applications on cortical surface functional magnetic resonance imaging (cs-fMRI) data and oceanic near-surface atmospheric data are conducted, showing that the proposed method has advantages over existing methods.
