Yield Spread Selection in Predicting Recession Probabilities: A Machine Learning Approach
This work addresses the problem of forecasting recessions for economists and policymakers, but it is incremental as it confirms existing methods rather than introducing new ones.
The study investigated whether machine learning could improve recession prediction by selecting optimal Treasury yield spread pairs, but found no significant improvement over the traditional 10-year--three-month spread due to estimation error.
The literature on using yield curves to forecast recessions customarily uses 10-year--three-month Treasury yield spread without verification on the pair selection. This study investigates whether the predictive ability of spread can be improved by letting a machine learning algorithm identify the best maturity pair and coefficients. Our comprehensive analysis shows that, despite the likelihood gain, the machine learning approach does not significantly improve prediction, owing to the estimation error. This is robust to the forecasting horizon, control variable, sample period, and oversampling of the recession observations. Our finding supports the use of the 10-year--three-month spread.