hh.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
A Universal Approach for Post-correcting Time Series Forecasts: Reducing Long-Term Errors in Multistep Scenarios
Halmstad University, School of Information Technology.ORCID iD: 0009-0002-2326-2401
Halmstad University, School of Information Technology.ORCID iD: 0009-0002-7613-9351
Halmstad University, School of Information Technology.ORCID iD: 0000-0001-8413-963x
2023 (English)In: Discovery Science: 26th International Conference, DS 2023, Porto, Portugal, October 9–11, 2023, Proceedings / [ed] Bifet A.; Lorena A.C.; Ribeiro R.P.; Gama J.; Abreu P.H., Cham: Springer, 2023, Vol. 14276 LNAI, p. 553-566Conference paper, Published paper (Refereed)
Abstract [en]

Time series forecasting is an important problem with various applications in different domains. Improving forecast performance has been the center of investigation in the last decades. Several research studies have shown that old statistical method, such as ARIMA, are still state-of-the-art in many domains and applications. However, one of the main limitations of these methods is their low performance in longer horizons in multistep scenarios. We attack this problem from an entirely new perspective. We propose a new universal post-correction approach that can be applied to fix the problematic forecasts of any forecasting model, including ARIMA. The idea is intuitive: We query the last window of observations plus the given forecast, searching for similar “shapes” in the history, and using the future shape of the nearest neighbor, we post-correct the estimates. To ensure that post-correction is adequate, we train a meta-model on the successfulness of post-corrections on the training set. Our experiments on three diverse time series datasets show that the proposed method effectively improves forecasts for 30 steps ahead and beyond. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Place, publisher, year, edition, pages
Cham: Springer, 2023. Vol. 14276 LNAI, p. 553-566
Series
Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349 ; 14276
Keywords [en]
Multi-steap Time series Forecasting, Post-correction
National Category
Probability Theory and Statistics
Identifiers
URN: urn:nbn:se:hh:diva-52062DOI: 10.1007/978-3-031-45275-8_37Scopus ID: 2-s2.0-85174303131Libris ID: 1jn47741z30jjmh0ISBN: 978-3-031-45274-1 (print)ISBN: 978-3-031-45275-8 (electronic)OAI: oai:DiVA.org:hh-52062DiVA, id: diva2:1812972
Conference
26th International Conference on Discovery Science, DS 2023, 9-11 October, 2023
Funder
Knowledge Foundation, #280033Available from: 2023-11-17 Created: 2023-11-17 Last updated: 2024-02-14Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textScopus

Authority records

Fanaee Tork, Hadi

Search in DiVA

By author/editor
Slepov, DennisKalinauskas, ArunasFanaee Tork, Hadi
By organisation
School of Information Technology
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 22 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf