Large and Small Order Flow as Equity Factors

Summary

This research examines whether large and small order money flows predict cross-sectional stock returns. It standardizes net inflow by trading value to form separate factors, then compares stocks grouped by factor strength. The reported results show a positive relationship for large-order flow and a negative one for small-order flow. The authors interpret the difference as large investors’ informational advantage and a crowding-out effect on smaller investors.

The study compares alternative scaling methods and reports the strongest results when net inflow is scaled by its absolute value. It also removes contemporaneous return effects through cross-sectional regressions: residual flow-strength factors show higher reported information coefficients and ratios, including within the CSI 300 and CSI 500 universes. A related residualization improves on a traditional reversal factor in the reported long-short results. These findings are backtest evidence as summarized in the document; the excerpt gives limited detail on sample construction, costs, implementation, or out-of-sample robustness, so it does not establish that the effects will persist in live trading.

Key ideas

• Large-order and small-order flow factors have opposite reported associations with subsequent stock returns.
• Scaling net inflow by its absolute value outperforms the other normalization methods described.
• Cross-sectional residualization is used to reduce the influence of contemporaneous returns on flow strength.
• Residual flow factors and residual reversal factors show improved reported information ratios.
• The excerpt does not provide enough detail to assess trading costs or out-of-sample durability.

Tags

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