Interpreting a Small-Cap Backtest and Momentum Factor Query

Summary

The post records a China A-share backtest using market capitalization ranks to select smaller companies, with filters for trading activity, positive trailing earnings, and positive price-to-earnings ratios. It reports that increasing the portfolio from 10 to 60 holdings and changing the rebalance interval affected performance, while a larger holding count and a longer interval reduced it. It also gives the Sharpe ratio formula and defines its return, risk-free rate, and volatility terms.

A second part asks why a SQL factor query does not work as intended. The example computes 1-, 30-, and 90-day returns, ranks returns and volume cross-sectionally, filters for stocks with strong recent returns, excludes special-treatment shares, and sorts by date and instrument. The post does not provide an answer, a complete backtest result, or enough detail to assess execution costs, survivorship bias, or out-of-sample robustness, so its performance observations should be treated as exploratory.

Key ideas

• The backtest selects smaller companies using market capitalization ranks and applies liquidity and fundamental filters.
• The author reports that holding count and rebalance interval changes affected the observed backtest performance.
• The example factor query computes trailing returns and ranks them across stocks before applying momentum-style filters.
• The post asks for help with the query but gives no resolution or evidence of robust out-of-sample performance.

Tags

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