Multi-step-ahead time series prediction using multiple-output support vector regression

• Published in: Neurocomputing (Amsterdam) 2014-04, Vol.129, p.482-493
• Main Authors: Bao, Yukun, Xiong, Tao, Hu, Zhongyi
• Format: Article
• Language: English
• Subjects: Accuracy; Applied sciences; Assessments; Computational efficiency; Computer science; control theory; systems; Data processing. List processing. Character string processing; Exact sciences and technology; Inference from stochastic processes; time series analysis; Mathematical models; Mathematics; Memory organisation. Data processing; Multi-step-ahead time series prediction; Multiple-input multiple-output (MIMO) strategy; Multiple-output support vector regression (M-SVR); Probability and statistics; Sciences and techniques of general use; Software; Statistics; Strategy; Support vector machines; Time series
• ISSN: 0925-2312; 1872-8286
• DOI: 10.1016/j.neucom.2013.09.010

Accurate time series prediction over long future horizons is challenging and of great interest to both practitioners and academics. As a well-known intelligent algorithm, the standard formulation of Support Vector Regression (SVR) could be taken for multi-step-ahead time series prediction, only relying either on iterated strategy or direct strategy. This study proposes a novel multiple-step-ahead time series prediction approach which employs multiple-output support vector regression (M-SVR) with multiple-input multiple-output (MIMO) prediction strategy. In addition, the rank of three leading prediction strategies with SVR is comparatively examined, providing practical implications on the selection of the prediction strategy for multi-step-ahead forecasting while taking SVR as modeling technique. The proposed approach is validated with the simulated and real datasets. The quantitative and comprehensive assessments are performed on the basis of the prediction accuracy and computational cost. The results indicate that (1) the M-SVR using MIMO strategy achieves the best accurate forecasts with accredited computational load, (2) the standard SVR using direct strategy achieves the second best accurate forecasts, but with the most expensive computational cost, and (3) the standard SVR using iterated strategy is the worst in terms of prediction accuracy, but with the least computational cost. •Propose M-SVR model with MIMO strategy for multi-step-ahead time series prediction.•Provide empirical evidence on three multi-step-ahead prediction strategies using SVR.•The M-SVR using MIMO strategy is the best with accredited computational load.•The computational load of standard SVR using direct strategy is extremely expensive.•The standard SVR using iterated strategy is best in terms of low computational load.