Forecast Accuracy Measures

If you study your past errors, at least you will learn how to win the last war.

—Paul Samuelson

What Is Forecast Error?

Forecast error is simply the difference between what we forecast and what actually happened. We make a forecast predicting what we think will happen. Then actually events happen and we can compare. This difference is forecast error. Forecast error is easily computed:

Forecast error: ET = DT – FT


ET = forecast error for period T

DT = actual value for period T

FT = forecast for period T

For example, we may have forecasted sales for January to be 300 units but we actually sold 320. Therefore, the forecast error for January would be 320 – 300 or 20. Notice that the number by itself doesn’t ...

Get Forecasting Fundamentals now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.