Title: Bayesian structural VAR models: An extended approach
Authors: Martin Bruns - University of East Anglia (United Kingdom)
Michele Piffer - Queen Mary, University of London (United Kingdom) [presenting]
Abstract: An approach is provided to derive the posterior distribution of SVAR models directly on the structural parameters. Our framework allows for restrictions not only on the contemporaneous relations among variables, but also on impact impulse responses. In so doing, we enrich the tool available for the researchers who aim to set identify structural VAR models. Applying the methodology to simulated data on the New Keynesian model, we find that our approach recovers the true responses more tightly than the popular indirect orthogonal reduced form approach. We then apply our procedure to the identification of fiscal shocks.