AI-Ranked China A-Shares with Trend Filters and Risk Controls
Summary
This strategy combines a restricted China A-share universe, model-based ranking, technical timing, and portfolio controls. It starts from a manually selected group of roughly 100 to 300 large-cap constituents associated with the CSI 150 universe. A predictive model ranks candidates for potential strength over the next several days; additional filters favor positive short- and medium-term returns relative to a benchmark, with stronger medium-term relative performance preferred.
Timing conditions use MACD relationships and the close relative to a moving average. The listed features include turnover, large-order net money flow, and a price-volume factor. A CSI 300 MACD crossover serves as a market risk trigger. The example holds five stocks, rotates on a short schedule, caps each position, and specifies profit-taking and loss-cutting thresholds. The post offers a strategy outline and says backtest results exist, but gives no performance figures, dates, or model validation details. It also does not explain the training process or address look-ahead bias, so claims of model quality cannot be evaluated from the description alone.
Key ideas
- The strategy uses a preselected pool of large China A-shares to narrow the model's search space.
- A predictive model ranks stocks, while relative-return filters favor recent leaders.
- MACD and a moving-average condition provide additional entry timing filters.
- Turnover, large-order money flow, and a price-volume feature inform ranking.
- Market-level MACD risk control, position limits, and explicit exit thresholds govern exposure.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.