SVR-FFS: A novel forward feature selection approach for high-frequency time series forecasting using support vector regression

作者：

Highlights：

• A novel approach for automatic time series forecasting is proposed.

• Support vector regression is adapted for forward feature selection.

• The proposal is successfully applied in energy load forecasting.

• Best predictive performance is achieved in experiments on these datasets.

摘要

•A novel approach for automatic time series forecasting is proposed.•Support vector regression is adapted for forward feature selection.•The proposal is successfully applied in energy load forecasting.•Best predictive performance is achieved in experiments on these datasets.

论文关键词：Support vector regression,Feature selection,Forecasting,Energy load forecasting,Automatic model specification

论文评审过程：Received 12 August 2019, Revised 5 July 2020, Accepted 5 July 2020, Available online 13 July 2020, Version of Record 23 July 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.113729