Discrete Probability Forecasts: What to expect when you are expecting a monetary policy decision

We apply discrete probability forecasts to the expectations of monetary policy rate changes, both in the United States and in the euro area. By using binomial trees from options theory, forecast distributions are derived from the instantaneous forward yield curve, based on interest rate swaps. We then use a non-randomised discrete probability forecast evaluation that confirms the presence of a systematic upward bias, consistent with the presence of a term premium. Consequently, we propose a bias-correction methodology to increase the accuracy of the density forecasts regarding monetary policy expectations. This research provides pivotal insights into understanding and improving predictive tools in monetary policy forecasting.