Research suggests that judgmental forecasts can also be improved by simply averaging the results of multiple independent forecasts.
Cognitive biases do much to explain the limitations of judgmental forecasts, but Tetlock's work suggests an additional explanation.
In cases where expert judgment is an appropriate method of forecasting, this article has identified a number of strategies for mitigating the weaknesses of judgmental forecasts, including documenting forecasts, not relying on a single forecaster, understanding cognitive biases and taking steps to mitigate them, and using some type of algorithm.
Stewart, "Improving Reliability of Judgmental Forecasts," in Armstrong, Principles of Forecasting.
Even in a less structured approach, getting more views for a judgmental forecast can be helpful--but beware of group forecasting.
As with most judgmental forecasts, the maximum projection horizon for the Greenbook forecast is not constant across vintages but varies from 6 to 10 quarters, depending on the forecast round.
The DSGE model and the BVAR produce continuous (and in very recent periods negative) interest rate forecasts, whereas the judgmental forecasts obviously factor in the discrete nature of interest rate setting and the zero nominal bound.
Here we adjust the DSGE model and BVAR forecasts using the realized future population growth numbers to make them comparable to announced GDP growth rates and judgmental forecasts, but we again note that this is an imperfect adjustment, which likely reduces the forecasting ability of the DSGE model and the BVAR.
It is not very surprising that judgmental forecasts fare better in capturing such regime switches.
Both the judgmental forecast and the DSGE model have an advantage over the purely statistical model, the BVAR, in that the people who produce the Greenbook and Blue Chip forecasts obviously know a lot more than seven time series, and the DSGE model was built to match the data that are being forecast.
The Greenbook forecast is a detailed judgmental forecast that until March 2010 (after which it became known as the Tealbook) was produced eight times a year by staff at the Board of Governors of the Federal Reserve System.