Section: New Results
Clustering for functional data
Participants : Julien Jacques, Cristian Preda.
In Jacques & Preda 2014 (CSDA), we propose a model-based clustering algorithm for multivariate functional data, based on multivariate functional principal components analysis. A review on clustering for functional data has also be published in Jacques & Preda 2014 (ADAC). Variable selection in high-dimensional regression Participants: Julie Hamon, Julien Jacques, Clarisse Dhaenens. In the context of genomic analysis, dealing with high-throughput genotyping data, we develop a genetic algorithm which looks for the best subset of variables (of given size) to predict some quantitative feature.