Within this tool, we track which agency is more accurate in terms of their forecasts. By comparing forecasts of different agencies across time against actual values (after the official figures for these indicators are released), we calculate the average error for each indicator and each forecasting horizon. The lower the average error, the better the forecast.

We are happy to provide the full methodology used in this tool. Please request the methodology by sending an email to: [email protected].

We fully understand that different agencies release their forecast figures at different times throughout the year. Therefore, some agencies work with more historical data than others before they generate their forecasts. That implies that these agencies are in a better position to forecast the future. We fully understand the limitation of our approach.

Another important point to keep in mind is that different agencies cover different set of indicators/countries/forecast periods. Additionally, we don’t have the same number of archived forecast releases for all agencies. Ideally, a true comparison of which agency is better in terms of forecasting is only possible under the exact conditions for all agencies. That is not realistic given different priorities and production schedule for different forecasting agencies.

Having identified these potential limitations, we still believe the tool is extremely useful as it could serve as a guide to which forecasting agency to use for specific indicators and countries.

Please review and let us know how we can improve the tool together.