Abstract
Statistical criteria for selecting a best forecasting method from a group of candidates have been proposed, studied, and implemented widely in forecasting software. Very well-known are information criteria, such as the AIC, which balance performance and complexity, and validation techniques, which examine forecasting performance in a holdout sample. So it's a breath of fresh air to have a distinctly new take on method selection, which is what Fotios and Enno are presenting here. They offer strong evidence that method selection can be improved by accounting for the representativeness of the forecasts. Copyright International Institute of Forecasters, 2022
Original language | English |
---|---|
Pages (from-to) | 5-12 |
Number of pages | 8 |
Journal | Foresight: the International Journal of Applied Forecasting |
Volume | 2022 |
Issue number | 65 |
Publication status | Published - 31 Dec 2022 |