Some authors argue that the delayed overshooting puzzle often found in the literature is an artifact of improper identification of monetary policy shocks, like Cholesky ordering. We investigate this claim by estimating the dynamic effect of monetary policy shocks on exchange rate using various identification schemes, where the data is generated by a small open economy DSGE model. We find that, on large sample, Cholesky type of restrictions perform comparably with model-consistent sign restrictions approach and do not produce the puzzle artificially. On short samples, however, Cholesky restrictions produce a more uncertain estimate for the shape of the exchange rate than sign restrictions.
Klodiana Istrefi and Balazs Vonnak
Classification JEL : E52, F41, C32
Keywords : Monetary Policy; Exchange Rate; DSGE; Vector Autoregressions; Cholesky Decomposition; Sign restrictions
Updated on: 06/12/2018 10:56