Research activity:
My research activity has mainly focused on graphical models from a methodological, computational and applied perspective. In this context, I have developed several Bayesian methodologies for structure learning (model selection) of Directed Acyclic Graphs (DAGs) both in Gaussian and categorical settings. More recently, I have applied graphical models to causal inference problems, and developed methods which combine structure learning and causal effect estimation. To this purpose, I co-authored the R package BCDAG https://cran.r-project.org/web/packages/BCDAG/index.html. For more information on my research you can refer to Section "Publications" and to the github repository https://github.com/FedeCastelletti which contains most of the codes implementing my proposed statistical methods.
Mail:
federico.castelletti [at] unicatt.it
Address:
UniversitĂ Cattolica del Sacro Cuore
Department of Statistical Sciences
Largo Gemelli 1, Milan