An economic forecast is an estimate of the future values of one or more economic variables, such as real gross domestic product (GDP), employment or price measures. There are many ways to make economic forecasts, and the resulting estimates are typically published in reports. A widely used method is econometric modeling, which applies an econometric algorithm to historical data inputs and assumptions. Other methods include consensus forecasts, simulated horizon analysis and supply-side forecasting.
In a world in which governments and individual economic actors—be they organizations or people—act on the basis of what they expect about the future, economic forecasts can have important real-world consequences. For example, a prediction that appears to be a wild overestimate of employment in 1992 might have had a significant impact on the U.S. presidential election that year.
For this reason, the measurement of the accuracy of a economic forecast is not a simple matter of applying once and for all some formula to assess prediction error. The estimated, realized values of many economic variables—such as employment and GDP—are often subject to multiple revisions.
Another challenge to measuring the accuracy of a economic forecast is that, unlike the law of gravity, the underlying dynamics of economic processes are not readily apparent. As a result, methodologies for making economic forecasts tend to focus on statistical characterizations of the variables that are being predicted rather than on theoretical explanations. It is therefore not surprising that, like the now-discredited methods that Galileo and Newton used to predict the movements of the planets, many of these methods produce good results even though they can be later seen as flawed.