Reproduce “FinReport: Explainable Stock Earnings Forecasting via News Factor Analyzing Model” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Xiangyu Li, Xinjie Shen, Yawen Zeng, Xiaofen Xing, Jin Xu
FinReport is an automated system developed by researchers from South China University of Technology and ByteDance AI Lab that democratizes sophisticated stock analysis for individual investors by integrating news factor analysis with traditional financial models. The system achieved an annualized return of 57.76% and a Sharpe ratio of 7.4043, outperforming market indices and baseline approaches during its evaluation.
Key insights
- Quantifying the impact of financial news through advanced NLP, specifically Semantic Role Labeling and Semantic Dependency Parsing, can significantly enhance stock return forecasting.
- Integrating news-derived factors into traditional multi-factor models, such as the Fama-French 5-factor model, improves predictive accuracy for stock movements.
- Automated systems can democratize access to sophisticated financial analysis by providing explainable reports for ordinary investors without requiring specialized knowledge.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.