EUVIMEDEUVIMED entdecken
Crossref

Rolling Forecast Model Selection through a Hybrid Multi-Metric Framework

Coralia Tanasuica (Zotic)

Applied Business: Issues & Solutions · 04.07.2026 · 2 · 71-76

DOI: 10.57005/ab.2026.2.10 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Business: Issues & Solutions, online erschienen am 04.07.2026. Verfasst von Coralia Tanasuica (Zotic). Die Arbeit nennt 11 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Retailers and producers worldwide need to be prepared to meet demand as effectively as possible to maximize profit, which means reducing or even eliminating out-of-stock situations while minimizing excess stock levels. Achieving highly accurate demand forecasts within a short time frame is therefore a key requirement. When dealing with multiple forecasting algorithms trained at the product level to learn sales patterns and demand behaviour, the complex part is to identify which algorithm best fits and forecasts each product’s behaviour. The present study proposes a structured, hybrid methodology for identifying the most suitable forecasting algorithm by combining rolling forecast model selection with a multi-horizon perspective, multiple evaluation metrics, and an adaptation of the Borda scoring method. The methodology is applied in a real operational context, using data provided by a large integrated company operating in both manufacturing and retail distribution within the construction materials sector in Eastern Europe.

Zitieren

APA

(Zotic), C. T. (2026). Rolling Forecast Model Selection through a Hybrid Multi-Metric Framework. Applied Business: Issues & Solutions, 2, 71-76. https://doi.org/10.57005/ab.2026.2.10

Vancouver

(Zotic) CT. Rolling Forecast Model Selection through a Hybrid Multi-Metric Framework. Applied Business: Issues & Solutions. 2026;2:71-76. doi:10.57005/ab.2026.2.10

BibTeX
@article{Zotic2026,
  author = {(Zotic), Coralia Tanasuica},
  title = {Rolling Forecast Model Selection through a Hybrid Multi-Metric Framework},
  journal = {Applied Business: Issues & Solutions},
  year = {2026},
  volume = {2},
  pages = {71--76},
  doi = {10.57005/ab.2026.2.10}
}
Kostenlos starten →