Identifying high-frequency shocks with Bayesian mixed-frequency VARs

We contribute to research on mixed-frequency regressions by introducing an innovative Bayesian approach. impose a Normal-inverse Wishart prior adding set of auxiliary dummies in estimating Mixed-Frequency VAR. Based this new “high-frequency” identification scheme, we illustrate our method identifying uncertainty shock for the U.S. economy. As main findings, document “temporal aggregation bias” when adopt common low-frequency model instead framework. The bias is amplified case large mismatching between high-frequency and business cycle variables.