Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Forecasting Four Business Cycle Phases Using Machine Learning: A Case Study of US and EuroZone

Elvys Linhares Pontes, Mohamed Benjannet, Raymond Yung

TL;DR

This paper tackles forecasting four business cycle phases (expansion, slowdown, recession, recovery) for the US and EuroZone using machine learning. It constructs an annotated, PCA-based index-driven dataset spanning 1970–2022, and compares Multinomial Logistic Regression, Support Vector Machines, and Multi-layer Perceptron, with MLR delivering the best performance (Top1/Top2: US 75%/92.14%, EZ 65.25%/84.74%). The study also benchmarks against a rule-based predictor and analyzes phase-wise and transition-phase difficulties, highlighting the practical potential and limitations of ML for macro regime classification. The findings suggest ML-based regime forecasting can inform economic and financial decision-making, while underscoring the need to account for regional heterogeneity and clearer phase borders in future work.

Abstract

Understanding the business cycle is crucial for building economic stability, guiding business planning, and informing investment decisions. The business cycle refers to the recurring pattern of expansion and contraction in economic activity over time. Economic analysis is inherently complex, incorporating a myriad of factors (such as macroeconomic indicators, political decisions). This complexity makes it challenging to fully account for all variables when determining the current state of the economy and predicting its future trajectory in the upcoming months. The objective of this study is to investigate the capacity of machine learning models in automatically analyzing the state of the economic, with the goal of forecasting business phases (expansion, slowdown, recession and recovery) in the United States and the EuroZone. We compared three different machine learning approaches to classify the phases of the business cycle, and among them, the Multinomial Logistic Regression (MLR) achieved the best results. Specifically, MLR got the best results by achieving the accuracy of 65.25% (Top1) and 84.74% (Top2) for the EuroZone and 75% (Top1) and 92.14% (Top2) for the United States. These results demonstrate the potential of machine learning techniques to predict business cycles accurately, which can aid in making informed decisions in the fields of economics and finance.

Forecasting Four Business Cycle Phases Using Machine Learning: A Case Study of US and EuroZone

TL;DR

This paper tackles forecasting four business cycle phases (expansion, slowdown, recession, recovery) for the US and EuroZone using machine learning. It constructs an annotated, PCA-based index-driven dataset spanning 1970–2022, and compares Multinomial Logistic Regression, Support Vector Machines, and Multi-layer Perceptron, with MLR delivering the best performance (Top1/Top2: US 75%/92.14%, EZ 65.25%/84.74%). The study also benchmarks against a rule-based predictor and analyzes phase-wise and transition-phase difficulties, highlighting the practical potential and limitations of ML for macro regime classification. The findings suggest ML-based regime forecasting can inform economic and financial decision-making, while underscoring the need to account for regional heterogeneity and clearer phase borders in future work.

Abstract

Understanding the business cycle is crucial for building economic stability, guiding business planning, and informing investment decisions. The business cycle refers to the recurring pattern of expansion and contraction in economic activity over time. Economic analysis is inherently complex, incorporating a myriad of factors (such as macroeconomic indicators, political decisions). This complexity makes it challenging to fully account for all variables when determining the current state of the economy and predicting its future trajectory in the upcoming months. The objective of this study is to investigate the capacity of machine learning models in automatically analyzing the state of the economic, with the goal of forecasting business phases (expansion, slowdown, recession and recovery) in the United States and the EuroZone. We compared three different machine learning approaches to classify the phases of the business cycle, and among them, the Multinomial Logistic Regression (MLR) achieved the best results. Specifically, MLR got the best results by achieving the accuracy of 65.25% (Top1) and 84.74% (Top2) for the EuroZone and 75% (Top1) and 92.14% (Top2) for the United States. These results demonstrate the potential of machine learning techniques to predict business cycles accurately, which can aid in making informed decisions in the fields of economics and finance.
Paper Structure (17 sections, 1 equation, 2 figures, 7 tables)

This paper contains 17 sections, 1 equation, 2 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Annotated dataset
  4. Machine learning approaches
  5. Macroeconomic data
  6. Machine learning methods
  7. Multinomial Logistic Regression
  8. Support vector machines
  9. Multi-layer Perceptron
  10. Experimental setup
  11. Evaluation metric
  12. Rule-based business cycle predictor
  13. Training procedure
  14. Experimental evaluation
  15. Conclusion
  16. ...and 2 more sections

Figures (2)

  • Figure 1: Historic of phases of business cycle on the United States (a) and EuroZone (b). Recovery is represented by '1', expansion by '2', slowdown by '3' and recession by '4'.
  • Figure 2: Slope calculation of linear regression of the last values (window size of 6) of a meta-parameter.