Building a Volume-Surge Stock Factor from Intraday Returns and Volatility

Summary

This Chinese equities research reproduction develops a cross-sectional factor from minute-level volume surges. A surge is defined when the minute-to-minute volume increase exceeds that stock’s daily average increase by one standard deviation. For each surge, the method measures the return in the surge minute and the volatility over that minute plus the following four minutes. Daily averages are compared with the cross-sectional average; their absolute distances are treated as measures of how far each stock’s response departs from a moderate level.

The distances are aggregated over a rolling 20-trading-day window using their mean and standard deviation, then standardized and combined into volatility and return components and a final “moderate risk-taking” factor. The article reports a 2013–2022 backtest on the broad Chinese A-share universe with neutralization, including factor and long-short statistics. These are reported results, not independent validation. The reproduction also notes that its sample code uses a small random stock subset for resource reasons, so scaling and implementation details matter when reproducing the full study.

Key ideas

• Volume surges are identified from unusually large minute-to-minute increases in each stock’s trading volume.
• The method measures both immediate returns at surge minutes and return volatility over the following four minutes.
• It defines moderate responses by distance from the daily cross-sectional mean, then aggregates those distances over 20 trading days.
• Standardized return and volatility components are combined into a stock-selection factor.
• Reported backtest statistics come from the reproduced study and do not establish robustness beyond its stated universe and period.

Tags

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