CONICET | Buscador de Institutos y Recursos Humanos

When dealing with multivariate data robust PCA, like classical PCA, searches for directions with maximal dispersion of the data projected on it. Instead of using the variance as a measure of dispersion, a robust scale estimator sn may be used in the maximization problem. In this paper, we review some of the proposed approaches to robust functional PCA including one which adapts the projection pursuit approach to the functional data setting.