Year: 2024
Type: article
Abstract: With the rapid development of financial markets, the linkage between the bond and stock markets has become increasingly evident. This paper leverages four deep learning models—XGBOOST, Random Forest, GRU, and LSTM—to predict stock prices using the S&P 500 dataset, aiming to explore the underlying mechanisms between stocks and bonds. Among these models, XGBOOST demonstrated the highest accuracy in predicting stock price trends, accurately tracking price volatility. The Random Forest model also displayed strong robustness. In contrast, LSTM and GRU were slightly less effective at capturing short-term sudden fluctuations in this experiment. Ultimately, this research provides a novel fintech perspective for risk monitoring and early warning, laying a foundation for future financial market prediction and risk management.
Author: Bokun Dong
Institution: Southwestern University of Finance and Economics
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4405592275