Chinese Stock Screen Using Price Amplitude, Ten-Day Returns, and KDJ

Summary

This stock screen combines daily amplitude above 1%, a positive ten-day return below 35%, and an increase in the K value of the KDJ oscillator. The accompanying explanation treats amplitude as a way to focus on active stocks, the bounded return as a filter for stocks that have risen without exceeding the stated cap, and KDJ growth as a signal of improving short-term direction. The Python example also applies additional filters on circulating market value and ranks candidates by a large-order flow measure, although the main prose does not include those conditions in its final summary of the selection logic.

The post provides no backtest or evidence that the criteria generate returns. It cautions that KDJ changes do not reliably predict future prices and that repeatedly using fixed rules may lead to overfitting. It suggests adding technical and fundamental measures or using machine learning, but does not specify a validation method, trading costs, or risk controls. The screen is therefore an illustrative rule set, not a demonstrated trading system.

Key ideas

• The stated screen selects stocks with amplitude above 1%, positive ten-day returns below 35%, and rising KDJ K values.
• The code example adds market capitalization filters and sorts by a large-order flow measure.
• The post provides no evidence that the screening rules have been backtested successfully.
• KDJ changes may not predict future prices, and fixed rules may become overfit.

Tags

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