Multi-relational Graph Diffusion Neural Network with Parallel Retention for Stock Trends Classification

ArXiv ID: 2401.05430 “View on arXiv”

Authors: Unknown

Abstract

Stock trend classification remains a fundamental yet challenging task, owing to the intricate time-evolving dynamics between and within stocks. To tackle these two challenges, we propose a graph-based representation learning approach aimed at predicting the future movements of multiple stocks. Initially, we model the complex time-varying relationships between stocks by generating dynamic multi-relational stock graphs. This is achieved through a novel edge generation algorithm that leverages information entropy and signal energy to quantify the intensity and directionality of inter-stock relations on each trading day. Then, we further refine these initial graphs through a stochastic multi-relational diffusion process, adaptively learning task-optimal edges. Subsequently, we implement a decoupled representation learning scheme with parallel retention to obtain the final graph representation. This strategy better captures the unique temporal features within individual stocks while also capturing the overall structure of the stock graph. Comprehensive experiments conducted on real-world datasets from two US markets (NASDAQ and NYSE) and one Chinese market (Shanghai Stock Exchange: SSE) validate the effectiveness of our method. Our approach consistently outperforms state-of-the-art baselines in forecasting next trading day stock trends across three test periods spanning seven years. Datasets and code have been released (https://github.com/pixelhero98/MGDPR).

Keywords: Graph Representation Learning, Multi-relational Graphs, Stock Trend Prediction, Information Entropy, Time-Series Forecasting, Equities

Complexity vs Empirical Score

  • Math Complexity: 7.5/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail
  • Why: The paper employs advanced graph theory, diffusion processes, and signal processing (entropy/energy formulas) with extensive real-world data across multiple markets, released code, and backtested baselines.
  flowchart TD
    A["Research Goal<br>Predict Next-Day Stock Trends<br>Modeling Time-Evolving Inter-Stock Dynamics"] --> B["Input Data<br>Real-world markets: NASDAQ, NYSE, SSE"]

    B --> C["Graph Construction<br>Dynamic Multi-Relational Graph Generation<br>using Info Entropy & Signal Energy"]

    C --> D["Adaptive Refinement<br>Stochastic Multi-Relational Diffusion Process<br>Learn task-optimal edges"]

    D --> E["Representation Learning<br>Decoupled Scheme with Parallel Retention<br>Capture temporal & structural features"]

    E --> F["Output & Findings<br>Stock Trend Classification<br>Consistent SOTA Outperformance over 7 Years"]