Optimal monetary policy using reinforcement learning

Arbeitspapier

This paper introduces a reinforcement learning based approach to compute optimal interest rate reaction functions in terms of fulfilling inflation and output gap targets. The method is generally flexible enough to incorporate restrictions like the zero lower bound, nonlinear economy structures or asymmetric preferences. We use quarterly U.S. data from1987:Q3-2007:Q2 to estimate (nonlinear) model transition equations, train optimal policies and perform counterfactual analyses to evaluate them, assuming that the transition equations remain unchanged. All of our resulting policy rules outperform other common rules as well as the actual federal funds rate. Given a neural network representation of the economy, our optimized nonlinear policy rules reduce the central bank's loss by over43 %. A DSGE model comparison exercise further indicates robustness of the optimized rules.

Angaben zum Objekt

• ISBN: 978-3-95729-861-4
• Sprache: Englisch

Verweise und Beziehungen

• Erschienen in: Series: Deutsche Bundesbank Discussion Paper ; No. 51/2021

Klassifikation und Themen

• Klassifikation: Wirtschaft ; Neural Networks and Related Topics ; Optimization Techniques; Programming Models; Dynamic Analysis ; Monetary Policy ; Central Banks and Their Policies
• Thema: Optimal Monetary Policy ; Reinforcement Learning ; Artificial Neural Network ; Machine Learning ; Reaction Function

Beteiligte, Orts- und Zeitangaben

• Ereignis: Geistige Schöpfung
• (wer): Hinterlang, Natascha ; Tänzer, Alina
• Ereignis: Veröffentlichung
• (wer): Deutsche Bundesbank
• (wo): Frankfurt a. M.
• (wann): 2021

Weitere Informationen

• Handle: http://hdl.handle.net/10419/248736
• Letzte Aktualisierung: 10.03.2025, 11:42 MEZ