Financial time series forecasting using rough sets with time-weighted rule voting
作者：
Highlights：
• We propose a trend prediction model for financial assets with a rough sets approach.
• For the rough set model an adaptive time-weighted rule voting method is proposed.
• We do experiments with real life stock data, the results are compared with SVM.
• We show that our model outperforms other models, including baseline buy and hold.
• Time-weighted rule voting stabilizes classification accuracy of the model.
摘要
•We propose a trend prediction model for financial assets with a rough sets approach.•For the rough set model an adaptive time-weighted rule voting method is proposed.•We do experiments with real life stock data, the results are compared with SVM.•We show that our model outperforms other models, including baseline buy and hold.•Time-weighted rule voting stabilizes classification accuracy of the model.
论文关键词：Decision systems,Rough sets,Financial time series prediction
论文评审过程：Received 3 June 2016, Revised 27 August 2016, Accepted 29 August 2016, Available online 2 September 2016, Version of Record 16 September 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.08.066
