Using Volume Weighted Support Vector Machines with walk forward testing and feature selection for the purpose of creating stock trading strategy
作者：
Highlights：
• Forecasting short-term stocks trends using Volume Weighted SVM.
• Robust techniques of feature selection used to enhance classifier accuracy.
• Experiments show that presented approach performs better than basic classifier.
• Significant improvement of the rate of return and maximum drawdown achieved.
• We designed and built system for walk-forward testing of proposed strategy.
摘要
•Forecasting short-term stocks trends using Volume Weighted SVM.•Robust techniques of feature selection used to enhance classifier accuracy.•Experiments show that presented approach performs better than basic classifier.•Significant improvement of the rate of return and maximum drawdown achieved.•We designed and built system for walk-forward testing of proposed strategy.
论文关键词：Support Vector Machines,Trend forecasting,Walk-forward testing,Stock trading
论文评审过程：Available online 22 October 2014.
论文官网地址：https://doi.org/10.1016/j.eswa.2014.10.001
