Filtering Chinese Growth and Science Board Stocks in Model Data

Summary

This brief BigQuant Q&A addresses whether to exclude stocks listed on China’s ChiNext and STAR Market boards when building an AI trading strategy. Its central recommendation is to apply the exclusion during both model training and prediction, so the model is trained and used on a consistent universe. The page also raises the question of how to filter these stocks in a trade backtest module, but the supplied text does not give the requested code or a step-by-step implementation; it points readers to a strategy source and video instead.

The rationale offered is that including those stocks in the training data may affect the model and reduce its accuracy. No measurements, backtest results, or technical explanation support that claim, so it should be treated as guidance rather than demonstrated evidence. The note is useful as a reminder to align universe filters across training and prediction, but readers need the linked strategy material or platform documentation for implementation details.

Key ideas

• Apply the same board exclusions to training data and prediction data.
• The page argues that unfiltered stocks may affect model accuracy, but provides no supporting results.
• The stated question about implementing the filter in a trade backtest is not answered with code in the supplied text.

Tags

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