Enhancing Financial Forecasting with a Hybrid LSTM-Graph Neural Network Model

Le résumé fourni par la source

Forecasting stock prices is inherently challenging due to the dynamic and intricate nature of financial data. This research introduces an innovative model that integrates Long Short-Term Memory (LSTM) networks with Graph Neural Networks (GNN), addressing limitations of standalone approaches. Unlike conventional methods that separately analyze temporal patterns or interrelationships among stocks, this hybrid model processes both dimensions concurrently. This dual capability significantly enhances prediction precision and establishes a fundamental stock market information to validate its effectiveness. By combining the strengths of LSTM in capturing temporal dependencies and GNN in modeling relationships between stocks, the hybrid model achieves notable improvements over traditional machine learning and individual models. Experimental results demonstrate superior performance, evidenced by reduced Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values, as well as higher R2 scores, underscoring its accuracy and reliability. This model offers promising applications in domains such as financial forecasting, algorithmic trading, and portfolio optimization. By addressing both the temporal and relational complexities of stock data, the proposed approach sets a new benchmark for predictive analytics in the financial sector. Its ability to provide more accurate forecasts not only supports better decision-making but also opens avenues for further advancements in predictive modeling and financial technology.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.