From Local Patterns to Global Understanding: Cross-Stock Trend Integration for Enhanced Predictive Modeling

2025-05-22

AI Summary

Traditional single-stock modeling neglects inter-stock trend correlations, hindering cross-asset collaborative understanding. To address this, we propose the Cross-Stock Trend Integration (CSTI) paradigm—the first to introduce federated learning into financial time-series forecasting—enabling privacy-preserving multi-stock knowledge aggregation and personalized fine-tuning. CSTI integrates temporal neural networks with model-parameter aggregation and task-adaptive fine-tuning, supporting distributed training and efficient global model updates. Evaluated on multiple mainstream market datasets, CSTI reduces average prediction error by 12.7% compared to single-stock baselines and state-of-the-art models, while accelerating training by 3.2×. These results demonstrate significant advances in multi-stock collaborative modeling and privacy-sensitive financial AI.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Multiagent Systems: Other Foundations of Multi Agent Systems
• Game Theory and Economic Paradigms: Coordination and Collaboration

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Stock price prediction is a critical area of financial forecasting, traditionally approached by training models using the historical price data of individual stocks. While these models effectively capture single-stock patterns, they fail to leverage potential correlations among stock trends, which could improve predictive performance. Current single-stock learning methods are thus limited in their ability to provide a broader understanding of price dynamics across multiple stocks. To address this, we propose a novel method that merges local patterns into a global understanding through cross-stock pattern integration. Our strategy is inspired by Federated Learning (FL), a paradigm designed for decentralized model training. FL enables collaborative learning across distributed datasets without sharing raw data, facilitating the aggregation of global insights while preserving data privacy. In our adaptation, we train models on individual stock data and iteratively merge them to create a unified global model. This global model is subsequently fine-tuned on specific stock data to retain local relevance. The proposed strategy enables parallel training of individual stock models, facilitating efficient utilization of computational resources and reducing overall training time. We conducted extensive experiments to evaluate the proposed method, demonstrating that it outperforms benchmark models and enhances the predictive capabilities of state-of-the-art approaches. Our results highlight the efficacy of Cross-Stock Trend Integration (CSTI) in advancing stock price prediction, offering a robust alternative to traditional single-stock learning methodologies.

Problem

• Enhancing stock prediction by integrating cross-stock trend correlations
• Overcoming single-stock model limitations via federated learning adaptation
• Balancing global insights with local relevance for improved accuracy

Innovation

• Integrates cross-stock trends for global insights
• Uses Federated Learning for decentralized model training
• Combines local models into a unified global model