B0629
Title: Analyzing data in complex 3D domains: Smoothing, semiparametric regression and functional principal component analysis
Authors: Eleonora Arnone - University of Turin (Italy) [presenting]
Letizia Clementi - Politecnico di Milano (Italy)
Laura Sangalli - Politecnico di Milano (Italy)
Abstract: A family of methods is introduced for the analysis of data observed at locations scattered in 3D domains, with possibly complicated shapes. The proposed family of methods includes smoothing, regression and functional principal component analysis for functional signals defined over (possibly non-convex) 3D domains, appropriately complying with the non-trivial shape of the domain. The common building block of the proposed methods is a nonparametric regression model with differential regularization. The asymptotic properties of the methods are derived and compared, through simulation studies, to the available alternatives for the analysis of data in 3D domains. An application is finally illustrated in a neurosciences study, with neuroimaging signals from functional magnetic resonance imaging, measuring neural activity in the grey matter, a non-convex volume with a highly complicated structure.
