Forecast With Judgment And Models

In this paper, the author proposes a model-consistent and parsimonious method of combining judgmental and model-based forecasts. The author suggests modeling the judgmental forecasts as optimal estimates of the variables of interest, made with a different, possibly more informative, information set. The author then shows how they can be accounted for in the framework of a linearized and solved DSGE model. The methodology the author proposes allows generating forecasts that are more accurate than the purely model-based ones, but that are still disciplined by the economic rigor of the model.