LocationRoom BZ E4.23, Universitätsplatz 1 - Piazza Università, 1, 39100 Bozen-Bolzano
Departments Press and Events
Contact Sabine Zanin
sabine.zanin@unibz.it
01 Dec 2022 12:30-13:30
Adaptive Subspace Shrinkage with Mixture Functional Horseshoe priors
research seminar
LocationRoom BZ E4.23, Universitätsplatz 1 - Piazza Università, 1, 39100 Bozen-Bolzano
Departments Press and Events
Contact Sabine Zanin
sabine.zanin@unibz.it
The seminar will be held in presence and online via teams: please click here
abstract: "Striking a balance between model complexity and parsimony is of crucial importance for producing accurate forecasts and structural inference. In this paper, we propose a non-parametric VAR model which relies on splines to approximate the unknown function between the response vector and its lags. This spline-based VAR is highly flexible and potentially overfits the data. To circumvent issues related to overfitting we introduce a prior which shrinks the non-parametric model towards a simpler model."