Automatic robust estimation for exponential smoothing: Perspectives from statistics and machine learning
作者：
Highlights：
• M-estimators, boosting and bagging are evaluated for exponential smoothing.
• M-estimators and machine learning approaches show improvements in accuracy.
• The Pseudo–Huber loss provides the best accuracy and bias reduction.
• Inverse boosting is comparable to M-estimators outperforming conventional boosting.
• Bagging achieves poor forecast bias compared to benchmark maximum likelihood.
摘要
•M-estimators, boosting and bagging are evaluated for exponential smoothing.•M-estimators and machine learning approaches show improvements in accuracy.•The Pseudo–Huber loss provides the best accuracy and bias reduction.•Inverse boosting is comparable to M-estimators outperforming conventional boosting.•Bagging achieves poor forecast bias compared to benchmark maximum likelihood.
论文关键词：Forecasting,Exponential smoothing,M-estimators,Boosting,Bagging
论文评审过程：Received 5 June 2019, Revised 25 April 2020, Accepted 4 June 2020, Available online 15 June 2020, Version of Record 3 July 2020.
论文官网地址：https://doi.org/10.1016/j.eswa.2020.113637
