AI-Ranked China A-Share Momentum Strategy with MACD Timing and Risk Controls

Summary

This strategy-sharing article describes an active China A-share approach that blends model-based stock ranking with technical timing. It starts from a manually selected China Securities 150-related universe, described as roughly 100–300 larger, liquid companies. An AI model ranks candidates, while MACD conditions, price above a 25-day average, and relative performance over short and longer windows are used to identify strong trends. The selected portfolio is concentrated, with five holdings and a five-day rotation schedule; the article also gives per-position sizing and explicit profit-taking and stop-loss thresholds.

The author characterizes the method as combining value-oriented selection with signal-based entry and trend following. Turnover and large-order money-flow features are included, and a CSI 300 MACD condition serves as a market risk trigger. The article mentions backtest results but supplies no figures or methodology in the provided text, so performance cannot be assessed here. Model details, training safeguards, transaction costs, slippage, and robustness across market regimes are not explained; the author’s stated rationale and thresholds should therefore be treated as a strategy description rather than validated evidence.

Key ideas

• The strategy ranks a restricted large-cap universe with an AI model, then applies technical entry filters.
• MACD, a 25-day moving average, and relative returns are used to screen for trend strength.
• Turnover and large-order money-flow features support liquidity and timing decisions.
• The described portfolio holds five stocks, rotates on a five-day schedule, and uses profit and loss exits.
• The text mentions backtesting but gives no results or methodology, and key model and trading assumptions remain unspecified.

Tags

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