Screening Stocks with Moving-Average Alignment and Turnover Filters

Summary

The article describes a Chinese stock screen combining three ideas: several moving averages clustered together, turnover within a specified band, and a 20-day moving average above the 120-day average. It interprets clustered averages as alignment across time horizons, moderate turnover as sufficient activity, and the faster average leading the slower one as a sign of stronger recent price direction. It includes a Python example intended to illustrate the filters.

The stated conditions change between the initial description and the final rule: the initial screen calls for at least five overlapping averages and turnover above 2% but below 9%, while the final version uses at least six and turnover above 3% but below 15%. The code also does not faithfully implement all of these conditions: its early return prevents later filtering from running, and its moving-average comparison loop does not test whether multiple averages overlap. No backtest or performance evidence is provided. The article flags market, technical-signal, and trading-cost risks, so the screen should be treated as an unvalidated candidate-selection idea.

Key ideas

• The proposed screen combines moving-average clustering, a turnover range, and a 20-day average above the 120-day average.
• The initial and final versions specify different minimum overlap counts and turnover bands.
• The sample code does not correctly apply the full set of described filters.
• The article reports no backtest and identifies market, indicator, and transaction-cost risks.

Tags

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