An adaptive feature selection schema using improved technical indicators for predicting stock price movements

作者：

Highlights：

• A wavelet denoising method is used to generate the improved technical indicators.

• A two-stage adaptive feature selection schema selects the optimal feature set.

• Multiple data sets are built based on 4 different stock markets.

• Improved technical feature selection improves predictive model performance.

摘要

•A wavelet denoising method is used to generate the improved technical indicators.•A two-stage adaptive feature selection schema selects the optimal feature set.•Multiple data sets are built based on 4 different stock markets.•Improved technical feature selection improves predictive model performance.

论文关键词：Stock price movement direction forecasting,Wavelet denoising,Feature selection,Time window

论文评审过程：Received 29 May 2021, Revised 16 March 2022, Accepted 17 March 2022, Available online 19 March 2022, Version of Record 29 March 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.116941