Skip to content
All library documents

Earnings Announcement Premium from Concentrated Trading Volume

Code Awesome Systematic Trading

Summary

This strategy ranks stocks monthly by the share of their trading volume occurring in recent earnings-announcement months. It uses a 48-month history and focuses on the latest 16 announcement months, then divides stocks into quintiles by the resulting volume-concentration ratio. Within the highest-concentration group, it goes long stocks whose announcement month is predicted from the prior year and short those not predicted to announce. Positions are weighted by market capitalization and rebalanced monthly.

The supplied implementation changes the original universe to the 1,000 most liquid US stocks listed on the NYSE, AMEX, or NASDAQ, and describes daily volume collection, monthly portfolio selection, and position execution. The document offers no reported backtest results or explanation of the economic mechanism behind the premium. Its code also contains implementation details that may affect whether it faithfully matches the written design, so the strategy needs careful review and validation before its results can be interpreted.

Key ideas

  • The strategy ranks stocks by trading volume concentrated in recent earnings-announcement months.
  • It forms a long-short position within the highest concentration group based on predicted announcement timing.
  • Portfolio weights are based on market capitalization, with monthly rebalancing.
  • The implementation uses a liquid US equity universe and provides no performance results.
  • The code should be checked against the written design before drawing conclusions.

Tags

EquitiesEvent-drivenFactor investingPortfolio construction

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