Skip to content
All library documents

Building and Reproducing an Intraday Volume Tide Factor

Article BigQuant

Summary

This document describes an equity factor based on intraday volume moving through a low-to-high-to-low cycle. It smooths volume with a nine-minute neighborhood, identifies a peak and the lowest-volume points before and after it, then measures price-change rates across the full cycle and its stronger and weaker halves. The final factor averages a 20-session measure of the stronger half with a stability measure for the weaker half. The proposed intuition is contrarian: sharp price moves during a volume cycle may reflect overreaction.

The author reports a reproduction over 2020–2024, comparing factor and portfolio metrics with the referenced research. Several reported measures are close, while the long-horizon return comparison differs materially. These results are presented as a reproduction, not independent validation; the document provides limited detail on data construction, costs, implementation choices, or robustness beyond the stated comparison.

Key ideas

  • The method defines an intraday volume cycle using smoothed volume, a peak, and surrounding troughs.
  • It computes price movement rates across the full cycle and its stronger and weaker halves.
  • The final factor combines a stronger-half measure with a weaker-half stability measure over a rolling 20-session window.
  • The document reports a 2020–2024 reproduction with broadly similar core metrics, alongside a large long-horizon return discrepancy.
  • The reported backtest does not establish robustness to transaction costs or alternative implementation choices.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.