Identification of Structural Vector Autoregressions by Stochastic Volatility

We propose to exploit stochastic volatility for statistical identification of Structural Vector Autoregressive models (SV-SVAR). We discuss full and partial identification of the model and develop efficient EM algorithms for Maximum Likelihood inference. Simulation evidence suggests that the SV-SVAR works well in identifying structural parameters also under misspecification of the variance process, particularly if compared to alternative heteroskedastic SVARs. We apply the model to study the interdependence between monetary policy and stock markets. Since shocks identified by heteroskedasticity may not be economically meaningful, we exploit the framework to test conventional exclusion restrictions as well as Proxy SVAR restrictions which are overidentifying in the heteroskedastic model.