A tale of two recession-derivative indicators

Area of Science:

• Economics
• Econometrics
• Financial Forecasting

Background:

• Recession-derivative indicators (RDIs) are crucial for business cycle prediction.
• Two primary RDIs exist: one for recessions starting at a specific future horizon, and another for recessions starting within a future period.

Purpose of the Study:

• To quantitatively compare the predictive performance of two distinct RDIs.
• To evaluate how forecast horizon, recession duration, and signal timing influence RDI effectiveness.
• To assess the utility of a semiannual RDI chronology.

Main Methods:

• Utilized daily yield spread as a predictor.
• Employed receiver operating characteristics (ROC) analysis for quantitative comparison.
• Analyzed data from 1962-2021, encompassing eight NBER-defined recessions.

Main Results:

• The second RDI (recession starting any time over a period) generally enhances predictor performance.
• The first RDI (recession starting at a specific horizon) can yield superior predictions in certain contexts.
• A semiannual RDI chronology demonstrated intermediate performance between the two primary RDIs.

Conclusions:

• The choice between RDIs depends on the specific needs of the forecast user and decision-making context.
• Understanding RDI nuances is vital for accurate business cycle forecasting.
• The study provides a quantitative framework for selecting appropriate RDIs.