Intraday Realized-Moment Factors for Short-Horizon Stock Selection

Summary

This research summary describes a short-horizon A-share selection factor built from intraday data. It calculates realized volatility, realized skewness, and realized kurtosis, then uses a regression model and the residual standard deviation from each period as the factor signal. The study evaluates whether the resulting measure distinguishes subsequent stock returns across different universes, including the full market and CSI 500 constituents.

The reported evidence covers a backtest beginning in 2007. The summary states that the factor showed return separation and monotonic returns across signal buckets, with negative average information coefficients in both universes. It also reports annualized portfolio returns, benchmark-relative performance, information ratios, and a maximum drawdown for the CSI 500 test. These are historical results from the summarized report, not a guarantee of future performance. The available text does not detail transaction costs, implementation assumptions, or robustness beyond the tested universes, and readers should consult the full study before interpreting the results.

Key ideas

• The factor uses intraday realized volatility, skewness, and kurtosis as regression inputs.\nResidual standard deviation from each period’s regression serves as the proposed stock-selection signal.\nThe summary reports return separation and monotonic bucket returns in the full market and CSI 500 universe.\nReported information coefficients are negative, so the signal’s direction matters when forming portfolios.\nThe evidence is historical backtesting, and the summary omits details on costs and implementation assumptions.

Tags

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