Testing Chinese Equity Selection Factors from Level 2 Trade Data
Summary
This report explores whether stock selection factors built from Level 2 transaction records can predict returns over a monthly horizon. It derives measures from trade size categories and trade direction, including the shares of trading represented by different order sizes and net buying ratios. The premise is that detailed intraday trades may contain information not captured by more commonly studied factor families.
The report says higher continuous-auction, medium-order, and small-order turnover shares were associated with stronger next-month returns, while lower very-large- and large-order shares were favored. Long-short tests for several turnover-share measures showed stronger results than net-buying-ratio measures; the latter generally had weak performance, with a late-session variant doing somewhat better. The turnover-share factors were also related to size, prior returns, and turnover. Cross-sectional regression was used to remove those exposures, after which the factors reportedly retained selection power. These are preliminary backtest findings; the supplied text gives no sample period, implementation details, or evidence on costs and out-of-sample robustness.
Key ideas
- Level 2 trade records support factors based on trade size and aggressor direction.
- Several trade-size turnover shares reportedly sorted stocks by subsequent monthly returns.
- Net-buying-ratio factors generally performed less well than turnover-share factors.
- Turnover-share factors had size, reversal, and turnover exposures that the report residualized through cross-sectional regression.
- The supplied summary describes preliminary backtests without enough detail to assess costs or out-of-sample stability.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.