Title: Identification, informational sufficiency and the role of monetary policy
Authors: Haroon Mumtaz - Queen Mary University of London (United Kingdom) [presenting]
Mirela Sorina Miescu - Lancaster University (United Kingdom)
Abstract: Informational sufficiency and a valid identification strategy are both necessary conditions to recover reliable impulse responses in structural vector autoregressive models (SVAR). We recommend the use of a Proxy Factor Augmented VAR model (FAVAR) which addresses the two conditions in a unified framework. We evaluate the performance of the Proxy FAVAR model versus a small scale Proxy SVAR in two Monte Carlo experiments. We show that the Proxy FAVAR model outperforms the Proxy SVAR in several cases of miss-specification as well as when the instrument is contaminated. In an empirical exercise, we examine the effects of monetary policy shocks on a large set of variables. We find that the impulse responses from a Proxy FAVAR model are considerably different from the ones delivered by a Proxy VAR, especially for real activity and prices variables. The results suggest that for an accurate evaluation of the effects of monetary policy shocks, it is crucial to complement a valid identification strategy with a large information set.