Detecting Stocks That Move Against the Market for Event-Driven Signals

Summary

This study turns unusual intraday stock behavior into a measurable event signal. It describes days when a stock repeatedly moves against the direction of the broader index, then uses correlation to screen for these cases. The resulting event samples are separated into stocks that rise against the market and those that fall against it, with the former reported to perform better. The study also forms a portfolio from the selected stocks and reports historical returns relative to broad Chinese equity benchmarks.

The evidence comes from a three-year historical backtest that identified 7,351 events, plus a portfolio period spanning April 2013 to February 2016. The report presents average excess returns over a 60-day holding period and cumulative and annualized portfolio results, but the available text does not explain screening thresholds, portfolio weighting, turnover, transaction costs, or whether the results were tested out of sample. The signal is proposed both as a trading factor and as a way to flag stocks for further fundamental research; the reported backtest should not be treated as evidence that the behavior will persist.

Key ideas

• The event signal identifies stocks that repeatedly move opposite to the index during a trading day.
• The screening method uses correlation between stock and index movements.
• The historical sample includes 7,351 detected events over three years.
• Stocks that rose against the market performed better than those that fell against it in the reported analysis.
• The report also tests a portfolio built from selected events, but the available summary omits implementation details such as costs and weighting.

Tags

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