Dipartimento di Scienze economico-sociali e matematico-statistiche

Descrizione del progetto

Building on the internationally recognized expertise of the research group, in this project we aim at developing the study of statistical properties of modern Bayesian nonparametric models. Bayesian nonparametrics is nowadays well established in several applications fields due to its internal consistency for modeling complex data without relying on stringent parametric assumptions. There is plenty of approaches that account for different types of dependence among observations, however the theoretical properties have somehow lagged behind. To fill this gap, we will investigate methodological and computational issues related to dependent nonparametric priors with emphasis on the large sample properties of posterior-based inference.