A1155
Title: Sparse causality networks through regularised regressions
Authors: Michele Costola - Ca' Foscari University of Venice (Italy) [presenting]
Mauro Bernardi - University of Padova (Italy)
Abstract: The aim is to propose a Bayesian approach to the problem of variable selection and shrinkage in high dimensional causal sparse regression models where the regularisation method is an extension of a previous LASSO in a Bayesian framework. The model allows us to extend the pairwise Granger (and quantile) causality in the network estimation by including a large number of institutions which improves the identification of the relationship and maintains at the same time the flexibility of the univariate framework. Furthermore, we obtain a weighted directed network since the adjacency matrix is built ``row by row'' using for each institutions the posterior inclusion probabilities of the other institutions in the network.
