Post-Filtering High-Frequency Short Factors in CSI 500 Enhancement

Summary

The document compares methods for using high-frequency factors with strong short effects to improve a CSI 500 index-enhancement strategy. It distinguishes screening out stocks before building the strategy from removing them afterward, and describes combining information either at the factor level or by combining short portfolios. The backtest summary finds that post-screening works across a wider range of short thresholds, suggesting greater parameter robustness than pre-screening.

At a 5% threshold, portfolio-level combination increases annualized excess return by more than 1%, while factor-level combination improves it by less than 1%. The document favors portfolio-level post-screening and reports that, with or without constituent constraints, thresholds from 5% to 10% raise annualized excess return by at least 0.9%. These are reported backtest results; the summary provides no detailed sample period, transaction-cost assumptions, or performance breakdown. It flags model specification, failure of statistical patterns, and liquidity as risks.

Key ideas

• The study compares removing stocks before strategy construction with filtering them afterward.
• Short signals can be combined by aggregating factors or by combining portfolios.
• Post-screening has a wider effective threshold range and is presented as more robust.
• Portfolio-level combination produces larger reported excess-return improvements than factor-level combination.
• The findings are backtest results and carry model, statistical stability, and liquidity risks.

Tags

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