AI-Ranked China 150 Stocks with Trend and Momentum Filters

Summary

This strategy combines model-based stock ranking with trend-following entry filters to seek short- and medium-term gains in Chinese equities. It starts from a manually selected pool of China Securities 150 constituents, narrowed to roughly 100–300 large, liquid stocks. An AI model ranks candidates, while MACD conditions and a price-above-25-day-average rule help time entries. Turnover, large-order net money flow, and a price-volume factor are used as ranking inputs. A target is defined around five-day relative returns, with an additional 30-day relative-strength percentile filter.

Portfolio controls include five holdings, a five-day rotation interval, a 10% per-stock allocation cap, a 60% profit-taking threshold, and a 5% loss exit. A MACD bearish crossover in the CSI 300 serves as a broad-market risk trigger. The author describes the approach and refers to backtest results, but the supplied text includes no performance figures or methodological validation. Model construction, data timing, transaction costs, and out-of-sample robustness are not detailed, so the reported design should not be treated as evidence of predictive edge.

Key ideas

• The approach blends AI ranking of a restricted large-cap stock universe with technical entry filters.
• MACD, a 25-day moving average, turnover, and large-order money flow inform selection and timing.
• Candidates are filtered for positive short- and medium-term relative performance.
• The portfolio uses five positions, a five-day rotation, and per-stock allocation and exit limits.
• The CSI 300 MACD signal is used as a broad-market risk control, but performance evidence is not provided in the text.

Tags

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