A combination method for interval forecasting of agricultural commodity futures prices
作者：
Highlights：
• Proposing an interval forecasting method for agricultural commodity futures prices.
• Extending the “linear and nonlinear” modeling framework for ITS forecasting.
• VECM and MSVR are integrated (abbreviated as VECM–MSVR).
• The experimental analysis is based on one-step-ahead and multi-step-ahead forecasts.
• VECM–MSVR is a promising method for interval forecasting in future markets.
摘要
•Proposing an interval forecasting method for agricultural commodity futures prices.•Extending the “linear and nonlinear” modeling framework for ITS forecasting.•VECM and MSVR are integrated (abbreviated as VECM–MSVR).•The experimental analysis is based on one-step-ahead and multi-step-ahead forecasts.•VECM–MSVR is a promising method for interval forecasting in future markets.
论文关键词：Interval-valued data,Agricultural commodity futures price forecasting,Vector error correction model (VECM),Multi-output support vector regression (MSVR),Econometric modeling
论文评审过程：Received 12 August 2014, Revised 18 November 2014, Accepted 5 January 2015, Available online 9 January 2015.
论文官网地址：https://doi.org/10.1016/j.knosys.2015.01.002
