A framework for economic demand forecast evaluation

Accurate demand forecasting is crucial for companies, since both overage and shortage of products can reduce company profits. Numerous statistical measures have been proposed to evaluate the accuracy of forecasts, but these measures tend to neglect the different economic impact of forecast errors for different items. Motivated by industry practitioners’ need to evaluate their forecasting processes and identify potentials for improvement, we develop a framework to evaluate demand forecasts. The framework consists of statistical measures and a scoring model and allows for an analysis of forecast accuracy on different levels of aggregation. A case study from a large retailing company shows that the framework is easy to implement in standard software and can provide relevant managerial insights.