Minute-Level Trade-Size Factors for Stock Selection and Reversal

Summary

This report summary describes ways to derive equity selection factors from the distribution and time patterns of individual trade amounts within each minute. Proposed statistics include quantiles, standard deviation, skewness, and kurtosis. It also discusses factors based on the relationship between trade amounts and price or volume measures, interpreting them as proxies for the influence and sentiment of relatively large traders. A further method refines a reversal factor by examining how reversal varies with trade size.

The summary reports historical tests with Rank IC and Rank ICIR statistics, returns, win rate, and comparisons across broad indices and industry portfolios. Results vary by factor and universe; the reported performance is strongest in the CSI 1000 constituent group among the broad indices discussed. These figures are summaries rather than independently verifiable evidence: the underlying report is referenced but not included. The document does not specify enough about sample period, transaction costs, portfolio construction, or robustness checks to establish that the reported performance would persist in live trading.

Key ideas

• Minute-level trade amounts can be summarized with distribution statistics to create stock selection factors.
• The report interprets the influence of relatively large trades as a possible source of factor signals.
• Correlations between trade amounts and price or volume measures are used to characterize trading intensity and sentiment.
• The proposed strong reversal factor examines reversal behavior in relation to trade size.
• Reported factor results differ across indices and industry portfolios, and the summary omits details needed to assess robustness and live-trading feasibility.

Tags

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