Title: Logratio approach to modeling of densities with application to multivariate functional principal component analysis
Authors: Karel Hron - Palacky University (Czech Republic) [presenting]
Alessandra Menafoglio - Politecnico di Milano (Italy)
Peter Filzmoser - Vienna University of Technology (Austria)
Abstract: A concise methodology has been developed since the early 1980s to deal with compositional data - i.e., multivariate data carrying only relative information - through the logratios of their parts. In parallel, the logratio approach to capture the specific features of continuous distributions (densities), known as the Bayes space methodology, is recently developed intensively as well. The aim is is to provide specific details to the continuous case focusing on the implications in case of multivariate functional principal component analysis. The theoretical developments are illustrated with a real-world case study.