Fathoming empirical forecasting competitions’ winners

Alroomi, Azzam; Karamatzanis, George; Nikolopoulos, Kostas; Tilba, Anna; Xiao, Shujun

Abstract

The M5 forecasting competition has provided strong empirical evidence that machine learning methods can outperform statistical methods: in essence, complex methods can be more accurate than simple ones. This result, be as it may, challenges the flagship empirical result that led the forecasting discipline for the last four decades: keep methods sophisticatedly simple. Nevertheless, this was a first, and thus we could argue this may not happen again. There has been a different winner in each forecasting competition. This inevitably raises the question: can a method win more than once (and should it be expected to)? Furthermore, we argue for the need to elaborate on the perks of competing methods, and what makes them winners?

Citation

Alroomi, A., Karamatzanis, G., Nikolopoulos, K., Tilba, A., & Xiao, S. (2022). Fathoming empirical forecasting competitions’ winners. International Journal of Forecasting, 38(4), 1519-1525. https://doi.org/10.1016/j.ijforecast.2022.03.010

• Journal Article Type: Article
• Acceptance Date: Mar 31, 2022
• Online Publication Date: Oct 5, 2022
• Publication Date: 2022-12
• Deposit Date: Apr 22, 2022
• Publicly Available Date: Oct 11, 2022
• Journal: International Journal of Forecasting
• Print ISSN: 0169-2070
• Publisher: Elsevier
• Peer Reviewed: Peer Reviewed
• Volume: 38
• Issue: 4
• Pages: 1519-1525
• DOI: https://doi.org/10.1016/j.ijforecast.2022.03.010
• Public URL: https://durham-repository.worktribe.com/output/1207807