Momentum-integrated Multi-task Stock Recommendation with Converge-based Optimization

ArXiv ID: 2509.10461 “View on arXiv”

Authors: Hao Wang, Jingshu Peng, Yanyan Shen, Xujia Li, Lei Chen

Abstract

Stock recommendation is critical in Fintech applications, which use price series and alternative information to estimate future stock performance. Although deep learning models are prevalent in stock recommendation systems, traditional time-series forecasting training often fails to capture stock trends and rankings simultaneously, which are essential consideration factors for investors. To tackle this issue, we introduce a Multi-Task Learning (MTL) framework for stock recommendation, \textbf{“M”}omentum-\textbf{“i”}ntegrated \textbf{“M”}ulti-task \textbf{“Stoc”}k \textbf{“R”}ecommendation with Converge-based Optimization (\textbf{“MiM-StocR”}). To improve the model’s ability to capture short-term trends, we novelly invoke a momentum line indicator in model training. To prioritize top-performing stocks and optimize investment allocation, we propose a list-wise ranking loss function called Adaptive-k ApproxNDCG. Moreover, due to the volatility and uncertainty of the stock market, existing MTL frameworks face overfitting issues when applied to stock time series. To mitigate this issue, we introduce the Converge-based Quad-Balancing (CQB) method. We conducted extensive experiments on three stock benchmarks: SEE50, CSI 100, and CSI 300. MiM-StocR outperforms state-of-the-art MTL baselines across both ranking and profitable evaluations.

Keywords: Stock Recommendation, Multi-Task Learning (MTL), Ranking Loss, Momentum Indicator, Portfolio Optimization, Equities

Complexity vs Empirical Score

• Math Complexity: 7.0/10
• Empirical Rigor: 8.5/10
• Quadrant: Holy Grail
• Why: The paper introduces novel mathematical components including a custom Adaptive-k ApproxNDCG ranking loss function and a Converge-based Quad-Balancing (CQB) optimization method, demonstrating high technical depth. Empirically, it is rigorously tested on three real-world stock benchmarks (SEE50, CSI 100, CSI 300) with profitability evaluations, ablation studies, and open-sourced code, making it highly implementation-ready.

  flowchart TD
    A["Research Goal:<br>Improve Stock Recommendation<br>via Multi-Task Learning"] --> B["Data Inputs:<br>Price Series & Alternative Data<br>(SEE50, CSI 100, CSI 300)"]
    
    B --> C["Key Methodology:<br>MiM-StocR Framework"]
    
    C --> D["Momentum Integration:<br>Captures Short-term Trends"]
    C --> E["Adaptive-k ApproxNDCG:<br>List-wise Ranking Loss"]
    C --> F["Converge-based<br>Quad-Balancing (CQB):<br>Mitigates Overfitting"]
    
    D --> G["Multi-Task Learning Process"]
    E --> G
    F --> G
    
    G --> H["Key Findings & Outcomes:<br>1. State-of-the-art Ranking<br>2. Enhanced Profitability<br>3. Robust MTL Performance"]
    
    style A fill:#e1f5e1
    style H fill:#ffe6e6