Uncertainty measures from partially rounded probabilistic forecast surveys

Although survey‐based point predictions have been found to outperform successful forecasting models, corresponding variance forecasts are frequently diagnosed as heavily distorted. Professional forecasters who report inconspicuously low ex ante variances often produce squared forecast errors that much larger on average. In this paper, we document the novel stylized fact misalignment is related rounding behavior of survey participants. Rounding may reflect some participants employ a rather judgmental approach opposed using formal model. We use distinct numerical accuracies panelists' reported probabilities way propose several alternative and easily implementable corrections (i) can be carried out in real time, is, before outcomes observed, (ii) deliver significantly improved match between post uncertainty. According our estimates, uncertainty about inflation, output growth unemployment U.S. Euro area higher after correcting for effect. The increase share nonrounded responses recent years also helps understand trajectory average during since financial sovereign debt crisis.