Large-Trade Weighted W-Split Reversal Factor

Summary

This document describes a cross-sectional equity reversal factor that separates recent returns according to trading activity. For each stock, it looks back over 20 sessions, calculates average transaction value per trade each day, ranks the sessions by that measure, and subtracts the summed returns of the lower-activity half from those of the higher-activity half. The resulting difference is the factor score.

The document reports that this approach was more stable than a conventional 20-day return reversal factor, with an information ratio of 2.51 and a monthly win rate of 74%. It also says that using higher transaction-value quantiles as the split threshold improved results, and attributes the reversal effect mainly to large trades. These claims are presented without details here about the sample, portfolio construction, costs, or out-of-sample validation, so the reported performance should not be treated as evidence of live profitability. The page also notes that its platform instructions and resources refer to an older version.

Key ideas

• The factor uses average transaction value per trade to rank a stock’s sessions over a 20-session lookback.
• It compares summed returns on the 10 higher-activity sessions with those on the 10 lower-activity sessions.
• The proposed factor score is the higher-activity return sum minus the lower-activity return sum.
• The document reports stronger results when the split emphasizes higher transaction-value quantiles.
• It interprets large trades as the main microstructure source of the reversal effect, though validation details are not supplied.

Tags

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