Predicting long-term stock movements with fused textual features of Chinese research reports

作者：

Highlights：

• Predicting long-term stock movements with fused textual features of research reports.

• Constructing a dataset composed of research reports and historical prices.

• Proposing a multi-module feature fusion method based on pre-trained language models.

• The basic information of stocks plays an important role in long-term forecasting.

摘要

•Predicting long-term stock movements with fused textual features of research reports.•Constructing a dataset composed of research reports and historical prices.•Proposing a multi-module feature fusion method based on pre-trained language models.•The basic information of stocks plays an important role in long-term forecasting.

论文关键词：Long-term stock prediction,Research report,Financial text mining,Textual feature,Pre-trained language model

论文评审过程：Received 17 April 2022, Revised 20 July 2022, Accepted 26 July 2022, Available online 4 August 2022, Version of Record 16 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118312