Evaluating Forecasts at Multiple Horizons: An Extension of the Diebold–Mariano Approach
Journal of Forecasting · 2026-08 · 45 (5) · 2379-2392
DOI: 10.1002/for.70150 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Forecasting, online erschienen am 22.03.2026. Verfasst von Andrew Grant und 2 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Forecast accuracy tests are fundamental tools for comparing competing predictive models. The widely used Diebold–Mariano (DM) test assesses whether differences in forecast errors are statistically significant. However, its standard form is limited to pairwise comparisons at a single forecast horizon. A number of solutions to this exist in the literature. Relative to these, this paper, based on a Mahalanobis approach, rather than approaches based on asymptotic normality. Our method incorporates cross‐covariances between errors and generalizes the DM framework to account for overlapping forecast windows and autocorrelation. The test provides a transparent alternative for forecasters and practitioners needing unified inference across temporal spans. We use simulations to assess conditions under which our test performs well, showing promising results in situations where big data are likely to be applicable.
Zitieren
APA
Grant, A., Mrazik, M., & Satchell, S. (2026). Evaluating Forecasts at Multiple Horizons: An Extension of the Diebold–Mariano Approach. Journal of Forecasting, 45(5), 2379-2392. https://doi.org/10.1002/for.70150
Vancouver
Grant A, Mrazik M, Satchell S. Evaluating Forecasts at Multiple Horizons: An Extension of the Diebold–Mariano Approach. Journal of Forecasting. 2026;45(5):2379-2392. doi:10.1002/for.70150
BibTeX
@article{Grant2026,
author = {Grant, Andrew and Mrazik, Matus and Satchell, Steve},
title = {Evaluating Forecasts at Multiple Horizons: An Extension of the Diebold–Mariano Approach},
journal = {Journal of Forecasting},
year = {2026},
volume = {45},
number = {5},
pages = {2379--2392},
doi = {10.1002/for.70150}
}
EUVIMED