Macroeconometric analysis using machine learning

2023 Fiscal Year Final Research Report

• Project/Area Number: 20H01482
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Basic Section 07030:Economic statistics-related
• Research Institution: The University of Tokyo
• Principal Investigator: Shintani Mototsugu 東京大学, 大学院経済学研究科(経済学部), 教授 (00252718)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 機械学習 / 時系列モデル / 経済予測 / ビッグデータ / マクロ政策評価

Outline of Final Research Achievements

We adopted machine learning methods, such as neural networks and Lasso, for macroeconomic analysis and constructed an economic forecasting model that accounts for nonlinearity. In this process, we utilized not only traditional macroeconomic data but also big data, such as microdata, survey data, and high-frequency data, as well as alternative data, such as text data. Additionally, on the top of standard time series models, we conducted policy evaluations using dynamic stochastic general equilibrium (DSGE) models that incorporate the optimizing behavior of economic agents, such as households and firms, based on macroeconomic theory.

Free Research Field

マクロ経済学

Academic Significance and Societal Importance of the Research Achievements

ビッグデータが利用可能な場合、機械学習の手法を活用することにより、マクロ経済変数の将来予測精度が高まることや、マクロ経済変動における非線形性の重要性が確認された。この結果と家計や企業等の経済主体の最適化行動を含んだマクロ経済理論モデルを組み合わせることで、より望ましい政策評価や因果推論が可能になることが示された。