Combining ROE, Auction Flow, Momentum, and Turnover Filters for Stock Selection

Summary

This Chinese-language post outlines a stock screen combining price range, profitability, opening-auction net buying, and technical conditions. Its initial criteria are a daily high-low range of at least one price unit, return on equity above 15% in each of five years, and positive net buying by major participants during the auction. The expanded version adds a KDJ condition with K above D, a close above the 10-day moving average, and turnover below 5%. It includes example expressions for a screening platform and Python-style pseudocode.

The post offers no backtest results or quantified evidence that the screen performs well. It flags uncertainty in historical-data-based selection, possible lag or inaccuracy in auction-flow measures, and poor suitability of ROE filters for some newer or persistently loss-making firms. The examples also leave implementation details to resolve: the narrative threshold for price range and the sample code's comparison may not match, and the Python ROE grouping expression needs checking to ensure it enforces the intended five-year condition per stock. The screen is a proposal requiring data validation and testing.

Key ideas

• The initial screen combines a minimum daily price range, five consecutive years of high ROE, and positive auction net buying.
• The expanded screen adds KDJ confirmation, a close above the 10-day average, and a turnover cap.
• The post supplies example screening expressions but reports no backtest evidence.
• Auction-flow data may lag or be inaccurate, and historical selection does not guarantee future results.
• The implementation examples should be checked for consistency with the stated thresholds and five-year ROE rule.

Tags

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