Identifying monetary policy shocks via heteroskedasticity: a Bayesian approach

In this paper we contribute to the literature on the identification of macroeconomic shocks by proposing a Bayesian SVAR with timevarying volatility of innovations that depend on a hidden Markov process, referred to as an MS-SVAR. With sufficient statistical information in the data, the distinct volatility regimes of the errors allow all the structural SVAR matrices and impulse response functions to be identified without the need for conventional a priori parameter restrictions. We give mathematical identification conditions and propose a flexible Gibbs sampling approach for the posterior inference on MS-SVAR parameters. The new methodology is applied to the US, euro area and Estonian macroeconomic series, where the effects of monetary policy and other shocks are examined.