A1217
Title: Bayesian dynamic tensor regression
Authors: Roberto Casarin - University Ca' Foscari of Venice (Italy)
Monica Billio - University of Venice (Italy)
Matteo Iacopini - Vrije Universiteit Amsterdam (Netherlands) [presenting]
Abstract: A new dynamic linear regression model is proposed for tensor variate response and covariates that encompasses univariate, multivariate (i.e. SUR, VAR, panel VAR) and matrix regression models as special cases. For dealing with the over-parametrization and over-fitting issues due to the curse of dimensionality, we exploit a suitable parametrization which enables to achieve both parameter parsimony and to incorporate sparsity effects. Inference is carried out in the Bayesian framework combined with Monte Carlo Markov Chain (MCMC). We show the eficiency of the MCMC procedure on simulated data, then we apply our methodology on macroeconomic and financial real datasets.
