A Bayesian method of combining judgmental and model-based density forecasts

This paper introduces a formal method of combining expert and model density forecasts when the sample of past forecasts is unavailable. It works directly with the expert forecast density and endogenously delivers weights for forecast combination, relying on probability rules only. The empirical part of the paper illustrates how the framework can be applied in forecasting US inflation by mixing density forecasts from an autoregressive model and the Survey of Professional Forecasters.

► We introduce a new method of combining expert and model-based density forecasts. ► It can be applied even if the previous history of forecasts is unavailable. ► It is able to work directly with expert's density forecast. ► The out-of-sample forecasting performance of the method is found to be promising.