Identification of monetary policy in SVAR models: a data-oriented perspective

In the literature using short-run timing restrictions to identify monetary policy shocks in vector-auto-regressions (VAR) there is a debate on whether (i) contemporaneous real activity and prices or (ii) only data typically observed with high frequency should be assumed to be in the information set of the central bank when the interest rate decision is taken. This paper applies graphical modeling theory, a data-based tool, in a small-scale VAR of the US economy to shed light on this issue. Results corroborate the second type of assumption.