We investigate the probability forecasting performance of a three-regime dynamic ordered probit model framework suitable to forecast recessions, low growth periods and accelerations for the U.S. and Japan. In a first step, we apply a non-parametric dating algorithm for the identification of...

This letter evaluates two types of forecasts from probit models that use the slope of the yield curve (“term spread”) to forecast NBER recessions, using an evaluation toolkit common in statistics and meteorology. The probit models provide excellent non-probabilistic yes/no predictions of...