Open Conference Systems, 50th Scientific meeting of the Italian Statistical Society

Font Size: 
Constructing priors for functional linear regression models
Maria Franco Villoria, Massimo Ventrucci, Haavard Rue

Last modified: 2018-05-28


In this work the functional linear regression model is considered as a varying coefficient model in a Bayesian setting. Elicitation of prior distributions that avoid overfitting is investigated and illustrated on a real case study following the recently introduced penalized complexity prior framework.