Boosting multi-step autoregressive forecasts

Souhaib Ben Taieb, Rob J. Hyndman

Research output: Chapter in Book/Report/Conference proceedingConference PaperOtherpeer-review

7 Citations (Scopus)


2014 Multi-step forecasts can be produced recursively by iterating a one-step model, or directly using a specific model for each horizon. Choosing between these two strategies is not an easy task since it involves a trade-off between bias and estimation variance over the forecast horizon. Using a nonlinear machine learning model makes the tradeoff even more difficult. To address this issue, we propose a new forecasting strategy which boosts traditional recursive linear forecasts with a direct strategy using a boosting autoregression procedure at each horizon. First, we investigate the performance of the proposed strategy in terms of bias and variance decomposition of the error using simulated time series. Then, we evaluate the proposed strategy on real-world time series from two forecasting competitions. Overall, we obtain excellent performance with respect to the standard forecasting strategies.

Original languageEnglish
Title of host publicationProceedings of the 31st International Conference on Machine Learning
EditorsEric P. Xing, Tony Jebara
PublisherInternational Machine Learning Society (IMLS)
Number of pages9
ISBN (Electronic)9781634393973
Publication statusPublished - 2014
EventInternational Conference on Machine Learning 2014 - Beijing International Convention Center (BICC), Beijing, China
Duration: 21 Jun 201426 Jun 2014
Conference number: 31st

Publication series

NameProceedings of Machine Learning Research
ISSN (Print)1938-7228


ConferenceInternational Conference on Machine Learning 2014
Abbreviated titleICML 2014
Internet address

Cite this