A0886
Title: Modeling cooperative fiscal policy in the Euro area using reinforcement learning and NMPC
Authors: Tato Khundadze - The New School (United States) [presenting]
Willi Semmler - New School for Social Research (United States)
Abstract: The purpose is to build on the experience of global shocks, such as the Eurozone crises of 2009-2012 and the economic crises resulting from COVID-19 starting in 2020. It aims to demonstrate the importance of cooperation in terms of monetary and fiscal policies during emergencies. The Euro area is chosen as the sample for testing the models presented in the paper. The case of the Euro area is crucial since its resilience heavily depends on cooperation between different actors within the region. The shocks affecting the nations within the European Union are asymmetric, and the responses to these shocks require coordination, considering the heterogeneous economic structures and levels of economic development. It is built on the previous study on cooperative and fiscal policies in the Euro Area, which uses nonlinear model predictive control (NMPC) to trace the path of key macroeconomic variables (inflation rate, interest rate, output gap, government gross debt, and price level) dynamics under non-cooperative and cooperative scenarios. The prior study is further extended by applying reinforcement learning to model the dynamics of the mentioned variables in a multi-agent environment.
