Combining Candlestick and Williams Shadows into an A-Share Selection Factor

Summary

This research summary tests stock-selection signals derived from candlestick upper and lower shadows, then compares them with shadow measures based on the Williams indicator. The reported results distinguish the two approaches: candlestick upper-shadow variability showed stronger selection ability than the candlestick lower-shadow factor, while the mean Williams lower-shadow measure performed relatively well and helped address the weaker lower-shadow result from candlesticks.

The authors combine those two measures into UBL and report tests on the full A-share universe from 2009 through April 2020. They give information-coefficient statistics, long-short portfolio results, monthly win rates, and maximum drawdown, and report that the combined factor retained selection ability after removing common style and industry effects. These are historical backtest findings from the summary, not proof of future performance. The supplied text does not explain implementation details, transaction costs, portfolio construction choices, or robustness across other markets and periods.

Key ideas

• The study turns candlestick upper and lower shadows into stock-selection factors.
• Candlestick upper-shadow variability performed better than the candlestick lower-shadow measure in the reported tests.
• A mean Williams lower-shadow factor showed stronger selection ability and complemented the candlestick signal.
• UBL combines the candlestick upper-shadow variability measure with the Williams lower-shadow mean.
• The reported A-share backtests cover 2009 through April 2020 and include style and industry adjustments.
• Historical results do not establish that the factor will perform in other periods or after implementation costs.

Tags

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