B0319
Title: The enlarged simplicial depth
Authors: Alicia Nieto-Reyes - Universidad de Cantabria (Spain) [presenting]
Giacomo Francisci - University of Trento (Italy)
Claudio Agostinelli - University of Trento (Italy)
Abstract: Most multivariate statistical data depth functions vanish right outside the convex hull of the support of the distribution with respect to which the depth is computed. This results in a challenge when we aim to distinguish among different points that are outside of the convex hull of the distribution support, with respect to which the depth is calculated. We give the first two definitions that overcome this issue for the simplicial depth function. We also present some corresponding depth estimators that do not vanish right outside the convex hull of the data. The properties of the definitions and of the respective estimators are illustrated, theoretically and by performing Monte Carlo simulations. Additionally, the importance of such proposals will be shown through the analysis of real data sets.
