At-Risk Transformation for U.S. Recession Prediction

2025-10-01

Working paper (Federal Reserve Bank of Philadelphia)

AI Summary

This study proposes a “risk-state” binarization approach to improve the accuracy of early U.S. recession forecasting. By converting continuous macroeconomic and financial variables into binary indicators that signal abnormally weak conditions, the method better aligns with the rare and discrete nature of recessions. Thresholds for binarization are data-driven, estimated from the training sample, and applied to a large panel of monthly predictors. The resulting binary features are then used in both linear and machine learning models for prediction. Empirical results demonstrate that this binarization significantly enhances out-of-sample forecasting performance, particularly at the onset of recessions, enabling even simple linear models to achieve predictive accuracy comparable to that of more complex machine learning methods.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Reasoning under Uncertainty: Relational Probabilistic Models
• Intelligent Robots: State Estimation

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

We propose a simple binarization of predictors, an"at-risk"transformation, as an alternative to the standard practice of using continuous, standardized variables in recession forecasting models. By converting predictors into indicators of unusually weak states based on a thresholding rule estimated from training data, we demonstrate their ability to capture the discrete nature of rare events such as U.S. recessions. Using a large panel of monthly U.S. macroeconomic and financial data, we show that binarized predictors consistently improve out-of-sample forecasting performance, often making linear models competitive with flexible machine learning methods, and that the gains are particularly pronounced around the onset of recessions.

Problem

• recession prediction
• rare events
• forecasting performance
• at-risk transformation
• binary predictors

Innovation

• at-risk transformation
• binary predictors
• recession forecasting
• thresholding
• out-of-sample prediction