Title: Bayesian semiparametric estimation of structural VAR models with stochastic volatility
Authors: Matteo Iacopini - Vrije Universiteit Amsterdam (Netherlands) [presenting]
Luca Rossini - University of Milan (Italy)
Abstract: The existing fully parametric Bayesian literature on structural VAR models with stochastic volatility (SVAR-SV) is extended by introducing an innovative Bayesian semiparametric framework to model high-dimensional time series of financial returns. A Bayesian nonparametric (BNP) approach based on a Dirichlet process mixture is used to flexibly model the returns distribution by also accounting for skewness and kurtosis, while the dynamics of each series volatility is modeled with a parametric structure. Our hierarchical prior overcomes overparametrization and over-fitting issues by clustering the coefficients into groups and shrinking the coefficients of each group toward a common location. An efficient Markov chain Monte Carlo sampling scheme is designed to perform inference in high-dimensional settings and provide a full characterization of parametric and distributional uncertainty. The proposed semiparametric approach is used to investigate returns predictability.