A digital graph in varying shades of blue overlays a background of buildings.

Brief

Forecast Accuracy and Forecaster Disagreement in the Survey of Professional Forecasters

Patrick Doelp

Research Associate

09 Jan ’24

Research Brief — This research brief continues the Philadelphia Fed’s longstanding tradition of examining the forecast accuracy of the SPF by testing the survey’s accuracy against a benchmark model, conditioning on forecaster disagreement.

The author of this brief examined forecast accuracy conditional on forecaster disagreement of nine variables within the SPF compared to a benchmark model using the Giacomini and White forecast accuracy test. The most general finding is that where there is a conditional difference in forecast accuracy between the SPF and the benchmark model, the SPF becomes more accurate than the benchmark model as forecaster disagreement rises in many of the examined cases. Additional evidence emphasizes the importance of using all four historical realizations of the data to test the forecasts against rather than relying on just a single realization for forecast accuracy tests.

View the Full Brief

Battle of the Forecasts: Mean vs. Median as the Survey of Professional Forecasters’ Consensus

Fatima Mboup &
Ardy Wurtzel

April 2021

We use the median forecast, because we've always assumed that the mean forecast is vulnerable to extreme individual responses. For the first time, we formally test our assumption in a statistically rigorous and comprehensive manner.

Macroeconomics

Brief

Battle of the Forecasts: Growth of the Median or Median of Growth as the SPF Consensus

January 2021

For the first time, we test which method produces the better consensus forecast, the Growth of the Median, or the Median of Growth.

Macroeconomics

Article

Forecast Disagreement in the Survey of Professional Forecasters

Keith Sill

Q2 2014

Business Review — To enact effective policies and spend resources efficiently, firms, policymakers, and markets need accurate economic forecasts. But even though economists generally work with similar models and data, their projections often range widely. To better understand why, Keith Sill explores what the evidence and theories say about how forecasters form their views.