Time series predictions using multi-scale support vector regressions

Authors:
Danian Zheng;Jiaxin Wang;Yannan Zhao
Affiliations:
Department of Computer Science and Technology, Tsinghua University, Beijing, China;Department of Computer Science and Technology, Tsinghua University, Beijing, China;Department of Computer Science and Technology, Tsinghua University, Beijing, China
Venue:
TAMC'06 Proceedings of the Third international conference on Theory and Applications of Models of Computation
Year:
2006

Citing 3
Cited 0

Least Squares Support Vector Machine Classifiers

Neural Processing Letters
An introduction to support Vector Machines: and other kernel-based learning methods

An introduction to support Vector Machines: and other kernel-based learning methods
Predicting Time Series with Support Vector Machines

ICANN '97 Proceedings of the 7th International Conference on Artificial Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Support vector regressions (SVR) have been applied to time series prediction recently and perform better than RBF networks. However, only one kernel scale is used in SVR. We implemented a multi scale support vector regression (MS-SVR), which has several different kernel scales, and tested it on two time series benchmarks: Mackey-Glass time series and Laser generated data. In both cases, MS-SVR improves the performance of SVR greatly: fewer support vectors and less prediction error.