Filtering ST-Designated Stocks from AI Strategy Data

Summary

This note explains how to exclude specially treated (“ST”) Chinese stocks from both the training and validation data flows in a visual AI stock-selection strategy. In the older workflow, it inserts a shared filtering module before missing-data handling in each flow. The newer approach applies an `st_status = 0` condition in the input feature list, filtering the universe at data selection.

The document gives implementation guidance but no performance results or comparison of the two methods. It identifies itself as an older implementation and points to a newer, simpler filter. The practical lesson is to apply the same eligibility rule to training and validation so the model’s data universe is consistent; the note does not discuss how ST status changes over time or how to prevent data leakage when historical labels are used.

Key ideas

• Exclude ST-designated stocks from both training and validation data.
• The older visual workflow adds a shared filter before missing-data processing.
• The newer workflow filters the input universe using the ST status field.
• The note describes implementation steps but provides no evidence about strategy performance.

Tags

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