Category Archives: Machine Learning

Identifying Monetary Policy Shocks: A Natural Language Approach

(with Thomas Drechsel)

We propose a novel method to identify monetary policy shocks. By applying naAruoba_Drechseltural language processing techniques to documents that economists at the Federal Reserve Board prepare for Federal Open Market Committee meetings, we capture the information set available to the committee at the time of policy decisions. Using machine learning techniques, we then predict changes in the target interest rate conditional on this information set, and obtain a measure of monetary policy shocks as the residual. An appealing feature of our procedure is that only a small fraction of interest rate changes is attributed to exogenous shocks. We find that the dynamic responses of macroeconomic variables to our identified shock measure are consistent with the theoretical consensus. We also demonstrate that our estimated shocks are not contaminated by the “Fed information effect.”

First Draft : March 2022

Paper

Most Recent Working Paper [March 2022]

Data Produced in the Paper (Monetary Policy Shock, Sentiments and FOMC Composition)