B1738
Title: Functional data analysis over multidimensional non-Euclidean domains
Authors: Laura Sangalli - Politecnico di Milano (Italy) [presenting]
Eleonora Arnone - University of Turin (Italy)
Letizia Clementi - Politecnico di Milano (Italy)
Alessandro Palummo - Politecnico di Milano (Austria)
Harold A Hernandez - Universidad Carlos III de Madrid (Spain)
M Carmen Aguilera-Morillo - Universitat Politecnica de Valencia (Spain)
Rosa Lillo - Universidad Carlos III de Madrid (Spain)
Abstract: An innovative class of methods are discussed for the analysis of functional data observed over multidimensional non-Euclidean domains, such as two-dimensional manifolds and non-convex volumes. The methods include functional principal component analysis and functional partial least squares and can handle sparse and partially observed functional data, as well as massive datasets. Challenging applications are illustrated to environmental and life sciences problems.