Graph-Based Learning for Stock Movement Prediction with Textual and Relational Data
This addresses the problem of improving stock movement prediction for financial analysts by considering stock correlations, though it appears incremental as it builds on existing sentiment extraction methods.
The paper tackles stock price prediction by incorporating both textual sentiment from financial news and relational information between correlated stocks, demonstrating better performance than benchmarks in accuracy tests and trading simulations on STOXX Europe 600 stocks.
Predicting stock prices from textual information is a challenging task due to the uncertainty of the market and the difficulty understanding the natural language from a machine's perspective. Previous researches focus mostly on sentiment extraction based on single news. However, the stocks on the financial market can be highly correlated, one news regarding one stock can quickly impact the prices of other stocks. To take this effect into account, we propose a new stock movement prediction framework: Multi-Graph Recurrent Network for Stock Forecasting (MGRN). This architecture allows to combine the textual sentiment from financial news and multiple relational information extracted from other financial data. Through an accuracy test and a trading simulation on the stocks in the STOXX Europe 600 index, we demonstrate a better performance from our model than other benchmarks.