Abstract
The authors criticize the exclusive use of forecast-error metrics to evaluate forecast methods, arguing for evaluations of forecast value, a construct that also incorporates considerations of the utility to the forecast user (influence on decision making), the cost of computation, and opportunity cost of understandability. They highlight evidence on forecast utility and cost from recent studies and explore the implications for forecast-method selection.
| Original language | English |
|---|---|
| Pages (from-to) | 36-45 |
| Number of pages | 10 |
| Journal | Foresight: the International Journal of Applied Forecasting |
| Volume | 63 |
| Publication status | Published - 31 Dec 2021 |