Forecasting Intraday Volume in Equity Markets with Machine Learning

2025-05-13

AI Summary

This paper addresses the challenge of intraday stock volume forecasting in high-frequency trading. Methodologically, it first systematically demonstrates the high predictability of intraday trading volume and innovatively introduces a common-component extraction technique to capture cross-stock shared factors. It then constructs a machine learning framework integrating XGBoost, LSTM, and ensemble models, trained via rolling windows and augmented with multi-source high-frequency features. Empirical results show substantial improvements in prediction accuracy over conventional benchmarks; backtesting a VWAP execution strategy yields an average reduction of 12.7% in transaction costs, confirming strong economic value. The key contributions are: (1) establishing the systematic predictability of intraday volume; (2) proposing a cross-stock common-factor modeling paradigm that enhances model robustness and generalizability; and (3) delivering a forecasting tool that achieves both statistical precision and practical efficacy for algorithmic trading.

Technology Category

• Machine Learning: Ensemble Methods
• Data Mining & Knowledge Management: Data Stream Mining
• Game Theory and Economic Paradigms: Mechanism Design

Application Category

• Web Mining and Content Analysis: Machine learning and data science for the Web
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: The sharing economy

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.

Problem

• Forecasting intraday trading volumes using machine learning
• Enhancing predictability with high-frequency predictors and ML models
• Assessing economic benefits of accurate volume forecasting via VWAP

Innovation

• Machine learning models for intraday volume forecasting
• High-frequency predictors enhance volume predictability
• VWAP strategies validate economic benefits