Dynamical analysis of financial stocks network: improving forecasting using network properties

ArXiv ID: 2408.11759 “View on arXiv”

Authors: Unknown

Abstract

Applying a network analysis to stock return correlations, we study the dynamical properties of the network and how they correlate with the market return, finding meaningful variables that partially capture the complex dynamical processes of stock interactions and the market structure. We then use the individual properties of stocks within the network along with the global ones, to find correlations with the future returns of individual S&P 500 stocks. Applying these properties as input variables for forecasting, we find a 50% improvement on the R2score in the prediction of stock returns on long time scales (per year), and 3% on short time scales (2 days), relative to baseline models without network variables.

Keywords: Network Analysis, Correlation Networks, Stock Return Prediction, S&P 500, Graph Theory, Equities

Complexity vs Empirical Score

• Math Complexity: 6.5/10
• Empirical Rigor: 7.5/10
• Quadrant: Holy Grail
• Why: The paper employs advanced mathematical concepts including complex network theory, multiple centrality measures, and dynamical systems analysis, requiring solid theoretical grounding. It demonstrates strong empirical rigor by applying these methods to real-world S&P 500 data across multiple time scales with reported quantitative performance improvements.

  flowchart TD
    A["Research Goal: Forecast S&P 500 returns<br>using network dynamics"] --> B
    subgraph B ["Data Preparation"]
        B1["Historical S&P 500<br>Price Data"]
        B2["Compute Correlations"]
    end
    B --> C["Construct Dynamic<br>Correlation Network"]
    C --> D["Calculate Features"]
    D --> E{"Prediction Tasks"}
    E --> F["Long-term<br>Annual Forecast"]
    E --> G["Short-term<br>2-Day Forecast"]
    F --> H
    G --> H
    subgraph H ["Model Training & Validation"]
        H1["Random Forest Regressor"]
        H2["Baseline vs<br>Network-Enhanced Models"]
    end
    H --> I
    subgraph I ["Key Findings"]
        I1["Annual R²: +50%<br>vs Baseline"]
        I2["2-Day R²: +3%<br>vs Baseline"]
        I3["Network Properties<br>capture market dynamics"]
    end