Cours/Séminaire
Notice
Lieu de réalisation
Paris
Langue :
Anglais
Crédits
François Baccelli (Publication), Giovanni Luca Torrisi (Intervention)
Détenteur des droits
Inria
Conditions d'utilisation
Droit commun de la propriété intellectuelle
Citer cette ressource :
Giovanni Luca Torrisi. Inria. (2023, 3 avril). Normal approximation of means of the Dirichlet-Ferguson measure , in DYOGENE/ERC NEMO 2023 : Seminar series. [Vidéo]. Canal-U. https://www.canal-u.tv/147606. (Consultée le 15 juin 2024)

Normal approximation of means of the Dirichlet-Ferguson measure

Réalisation : 3 avril 2023 - Mise en ligne : 3 avril 2023
  • document 1 document 2 document 3
  • niveau 1 niveau 2 niveau 3
Descriptif

The Dirichlet-Ferguson measure is a cornerstone in Bayesian nonparametrics and the study of the distributional properties of expectations with respect to (w.r.t.) such a measure is a line of research initiated in the seventies and still very active. We provide explicit upper bounds for (i) the Wasserstein distance between the law of a random mean w.r.t. the Dirichlet-Ferguson measure and the standard Gaussian distribution (ii) the d_2 and the convex distances between the law of a random vector, whose components are random means w.r.t. the Dirichlet-Ferguson measure, and a multivariate Gaussian distribution. As application of the general bounds, we study the normal approximation of linear transformations of random vectors distributed according to the Dirichlet law.

Based on a joint work with Ian Flint, University of Melbourne.

Intervention

Dans la même collection

Sur le même thème