Forecasting Intraday Volume in Equity Markets with Machine Learning

TL;DR

This paper demonstrates that machine learning models can effectively forecast intraday stock trading volumes, offering significant economic benefits for high-frequency trading strategies by leveraging numerous predictors and commonality effects.

Contribution

The study introduces a suite of machine learning models tailored for intraday volume prediction, filling a gap in flexible forecasting methods for high-frequency trading environments.

Findings

Intraday trading volume is highly predictable with ML models.

Accurate volume forecasts improve VWAP trading strategies.

Machine learning enhances predictability by incorporating multiple predictors.

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.