AI-Ranked China A-Shares with Momentum Timing and Portfolio Risk Rules

Summary

This strategy-sharing article describes an enhanced China Securities 150 equity approach that blends model-based stock ranking with technical timing. The universe is manually narrowed to roughly 100–300 large, liquid constituent-style stocks. An AI model ranks candidates for expected strength over the next five days, while MACD conditions and a close above the 25-day average act as entry filters. Additional features include turnover, large-order net flows, and five- and 30-day benchmark-relative returns; the relative-strength filter selects stocks with positive recent excess returns and a high 30-day rank.

Risk rules use a CSI 300 MACD signal as a market-level trigger, plus stated take-profit and stop-loss thresholds. The portfolio holds five names, rotates on a five-day schedule, and caps each position at 10% of capital. The article offers implementation details but no readable backtest statistics or evidence to assess performance. It also labels the platform resources as outdated, and its claims about model precision and factor behavior are not independently substantiated.

Key ideas

• The approach combines AI ranking of a restricted large-cap universe with technical entry timing.
• MACD, a 25-day moving average, turnover, and large-order net flows contribute to selection or timing.
• Relative performance over five and 30 days is used to favor stronger stocks.
• The portfolio uses five holdings, five-day rotation, a 10% per-stock allocation cap, and explicit profit and loss exits.
• The article provides strategy rules but no usable backtest results, and notes that its platform resources are outdated.

Tags

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