Volume-Ratio Stock Selection and Exit Rules in a BigQuant Backtest

Summary

This BigQuant example builds a daily Chinese equity strategy using a filtered stock universe and recent volume ratios. It screens stocks in the CSI 1000 universe across several board segments, excludes suspended and special-treatment stocks, and applies conditions to lagged volume ratios and price relative to a historical quantile. A scoring module ranks candidates, with the portfolio configured to hold five positions. The trading logic initially allocates across the top-ranked names, then considers profit-taking, loss exits, drawdown exits, and replacement using the current ranking.

The document’s title asks why a backtest trades only a few times and fails to trigger stop rules, but the body mainly reproduces code and does not diagnose that behavior or provide backtest results. The example’s exit logic depends on available daily data and price fields; its dynamic drawdown reference uses the position’s last sale price, which may not represent a running peak. The code also uses separate branches for exits and replacements, so its behavior should be checked against the platform’s order timing and position semantics before interpreting results.

Key ideas

• The example ranks Chinese stocks using lagged volume ratios and a price filter.
• The portfolio configuration targets five holdings and initially divides available cash among selected names.
• The trading logic includes profit-taking, loss exits, and a drawdown condition.
• The article does not explain why the reported backtest stops trading or why the exits fail to trigger.
• The drawdown reference and order timing may affect whether the exit logic behaves as intended.

Tags

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