Title: Identifying shocks in structural VAR models via heteroskedasticity: A Bayesian approach
Authors: Dmitry Kulikov - Eesti Pank (Estonia) [presenting]
Aleksei Netsunajev - Tallinn University of Technology (Estonia)
Abstract: A contribution to the literature on statistical identification of macroeconomic shocks is made by proposing a Bayesian VAR with time-varying volatility of the residuals that depends on a hidden Markov process, referred to as an MS-SVAR. With sufficient statistical information in the data and certain identifying conditions on the variance-covariance structure of the innovations, distinct volatility regimes of the reduced form residuals enable all structural SVAR matrices and impulse response functions to be estimated without the need for conventional prior identifying restrictions. We give mathematical identification conditions and propose a novel combination of the Gibbs sampler with a Bayesian clustering of impulse responses for the posterior inference on the MS-SVAR parameters. The new methodology is applied to the US data on output, inflation, real money and policy rates, where we demonstrate that the effects of two real and two nominal shocks are clearly identified by the new methodology.