Reproducing an Overnight Return Factor for Chinese Stock Selection

Summary

This note describes a reproduction of a sell-side factor based on overnight stock returns, intended to capture information advantages held by informed traders. It says the research defines the factor in stages, from a traditional overnight-return measure to a revised version that removes systematic returns; the implementation focuses directly on the final form. The author reports that the reproduced factor return curve looked similar to the report’s result and says a further stock-ranking backtest informed practical suggestions, but provides no detailed performance figures or statistical tests.

The reproduction differs from the source in rebalance timing, transaction-cost and price-adjustment assumptions, slippage, and the index used to estimate systematic returns. The author proposes excluding ST-designated stocks because their limited liquidity may create high market impact, combining the factor with momentum or reversal signals to address falling trends, and reversing the factor’s sign to align ranking direction with the trading setup. These are implementation observations rather than proof of robustness; the differences from the original study and the limited reported evidence constrain direct comparison and live-trading conclusions.

Key ideas

• The factor seeks to capture information advantages through overnight stock returns after removing systematic returns.
• The reproduction uses a regular trading-day rebalance cycle rather than month-end timing.
• Its costs, adjusted prices, slippage, and benchmark index differ from the original report.
• The author reports similar return-curve behavior but gives no detailed performance statistics.
• The note suggests liquidity exclusions and combining the signal with momentum or reversal measures.

Tags

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