Adaptive Factor Scoring for Chinese Stock Reversal Trades

Summary

The article describes selecting Chinese “monster stock” reversal candidates with a changing, multi-factor score. It examines recent profitable picks by industry and a normalized factor, then uses those observations to infer which sectors and factor ranges have recently been associated with positive outcomes. The author says the model trains on profitable stocks from the prior five days, scores the current candidate pool, and buys the highest-ranked names near the close. The factor example is illustrative; the document does not identify the actual factor or provide a reproducible scoring formula.

The author also gives discretionary trading rules: focus on stocks within an emotion-driven cycle, usually at five limit-up days or fewer; enter reversal candidates near the close; and exit the next afternoon if a stock is not limit-up or no longer appears in the candidate pool. The evidence is a brief account of recent picks and sector counts, without a complete sample, benchmark, risk-adjusted results, or out-of-sample test. The short training window may be noisy, and no controls for transaction costs, selection bias, or losses are described.

Key ideas

• The method scores candidate stocks using multiple factors rather than fixed thresholds.
• Recent profitable stocks inform sector preferences and whether factors favor higher, lower, or bounded values.
• The described training set consists of profitable stocks from the preceding five days.
• The author proposes buying top-ranked reversal candidates near the close and reviewing them the next afternoon.
• The article provides no full backtest or evidence that the short-window scoring method generalizes.

Tags

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