TY - JOUR

T1 - A comparison of AdaBoost algorithms for time series forecast combination

AU - Barrow, Devon Kennard

AU - Crone, Sven Friedrich Werner Manfred

N1 - This is the author’s version of a work that was accepted for publication in International Journal of Forecasting. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in International Journal of Forecasting, 32, 4, 2016 DOI: 10.1016/j.ijforecast.2016.01.006

PY - 2016/10/1

Y1 - 2016/10/1

N2 - Recently, combination algorithms from machine learning classification have been extended to time series regression, most notably seven variants of the popular AdaBoost algorithm. Despite their theoretical promise their empirical accuracy in forecasting has not yet been assessed, either against each other or against any established approaches of forecast combination, model selection, or statistical benchmark algorithms. Also, none of the algorithms have been assessed on a representative set of empirical data, using only few synthetic time series. We remedy this omission by conducting a rigorous empirical evaluation using a representative set of 111 industry time series and a valid and reliable experimental design. We develop a full-factorial design over derived Boosting meta-parameters, creating 42 novel Boosting variants, and create a further 47 novel Boosting variants using research insights from forecast combination. Experiments show that only few Boosting meta-parameters increase accuracy, while meta-parameters derived from forecast combination research outperform others.

AB - Recently, combination algorithms from machine learning classification have been extended to time series regression, most notably seven variants of the popular AdaBoost algorithm. Despite their theoretical promise their empirical accuracy in forecasting has not yet been assessed, either against each other or against any established approaches of forecast combination, model selection, or statistical benchmark algorithms. Also, none of the algorithms have been assessed on a representative set of empirical data, using only few synthetic time series. We remedy this omission by conducting a rigorous empirical evaluation using a representative set of 111 industry time series and a valid and reliable experimental design. We develop a full-factorial design over derived Boosting meta-parameters, creating 42 novel Boosting variants, and create a further 47 novel Boosting variants using research insights from forecast combination. Experiments show that only few Boosting meta-parameters increase accuracy, while meta-parameters derived from forecast combination research outperform others.

KW - Forecasting

KW - Time series

KW - Boosting

KW - Ensemble

KW - Model combination

KW - Neural networks

U2 - 10.1016/j.ijforecast.2016.01.006

DO - 10.1016/j.ijforecast.2016.01.006

M3 - Journal article

VL - 32

SP - 1103

EP - 1119

JO - International Journal of Forecasting

JF - International Journal of Forecasting

SN - 0169-2070

IS - 4

ER -