Identification Of Monetary Policy In SVAR Models: A Data-oriented Perspective
This paper applies graphical modeling theory to recover identifying restrictions for the analysis of monetary policy shocks in a VAR of the US economy. Results are in line with the view that only high-frequency data should be assumed to be in the information set of the monetary authority when the interest rate decision is taken. The literature employing Vector-AutoRegressions (VAR) to identify and estimate the effects of monetary policy shocks typically distinguish among three sets of variables: the information set, i.e. the set of variables known to the monetary authorities when the policy decision is taken; the policy instrument; the set of variables the value of which is known only after the policy is set.