Earnings Announcement Reversal Strategy: Long Losers, Short Winners

Summary

The strategy described in the code sorts stocks around earnings announcements by their returns from four to two trading days before the event. The underlying research description first divides stocks by firm size, then sorts the largest size group into return quintiles. It buys recent losers and shorts recent winners for a three-day holding window spanning the day before, day of, and day after the announcement, with equal weighting. The implementation uses earnings-date data, daily price histories, and a trade manager with separate long and short capacity limits.

The code provides a backtest implementation outline, including a custom fee model, leverage setting, universe selection, position sizing, and scheduled rebalancing. It does not report backtest returns or demonstrate that the effect persists. The implementation’s universe is limited to symbols in its supplied earnings data, and its available description does not establish out-of-sample robustness or address all practical concerns such as short availability, market impact, or data quality.

Key ideas

• The signal ranks pre-announcement returns and takes the opposite side of recent extremes.
• The research description first groups stocks by firm size and then sorts the largest group by returns.
• Positions are equally weighted and held across a three-day earnings window.
• The code includes fees, leverage, and limits on concurrent long and short positions.
• No performance results or evidence of robustness are provided.

Tags

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