Forecast combination for U.S. recessions with real-time data

• We use probability forecast combination to forecast the U.S. business cycle.
• We examine the coincident indicators and yield curve models’ forecasting performance.
• We use dynamics and real-time data revisions.
• Combining recession probability forecasts of both models improve forecast accuracy.
• Evidence favours model combinations over combining regressors into one model.

This paper proposes the use of forecast combination to improve predictive accuracy in forecasting the U.S. business cycle index, as published by the Business Cycle Dating Committee of the NBER. It focuses on one-step ahead out-of-sample monthly forecast utilising the well-established coincident indicators and yield curve models, allowing for dynamics and real-time data revisions. Forecast combinations use log-score and quadratic-score based weights, which change over time. This paper finds that forecast accuracy improves when combining the probability forecasts of both the coincident indicators model and the yield curve model, compared to each model's own forecasting performance.