Forecasting Intraday Volume in Equity Markets with Machine Learning

ArXiv ID: 2505.08180 “View on arXiv”

Authors: Mihai Cucuringu, Kang Li, Chao Zhang

Abstract

This study focuses on forecasting intraday trading volumes, a crucial component for portfolio implementation, especially in high-frequency (HF) trading environments. Given the current scarcity of flexible methods in this area, we employ a suite of machine learning (ML) models enriched with numerous HF predictors to enhance the predictability of intraday trading volumes. Our findings reveal that intraday stock trading volume is highly predictable, especially with ML and considering commonality. Additionally, we assess the economic benefits of accurate volume forecasting through Volume Weighted Average Price (VWAP) strategies. The results demonstrate that precise intraday forecasting offers substantial advantages, providing valuable insights for traders to optimize their strategies.

Keywords: Intraday Volume Forecasting, Machine Learning, Volume Weighted Average Price (VWAP), High-Frequency Predictors, Commonality, Equities

Complexity vs Empirical Score

• Math Complexity: 6.0/10
• Empirical Rigor: 7.5/10
• Quadrant: Holy Grail
• Why: The paper employs advanced machine learning models and statistical decompositions (CMEM, state-space models) indicating moderate-to-high mathematical complexity, while its use of high-frequency limit order book data, specific forecasting backtests, and economic evaluation via VWAP strategies demonstrates substantial empirical rigor.

  flowchart TD
    A["Research Goal<br>Forecast Intraday Trading Volume"] --> B["Data Collection<br>High-Frequency Equity Data"]
    B --> C["Feature Engineering<br>High-Frequency Predictors + Commonality"]
    C --> D{"Machine Learning Models<br>Ensemble Approach"}
    D --> E["Model Training & Validation"]
    E --> F["Forecast Generation"]
    F --> G["Backtesting<br>VWAP Strategy Simulation"]
    G --> H["Key Outcomes<br>High Predictability + Economic Value"]