A Beginner Workflow for Building and Testing Stock Selection Rules

Summary

This Chinese-language article introduces a workflow for building custom stock-selection strategies on the older BigQuant platform. It contrasts platform-generated AI strategies, which learn from selected factors but may be difficult to interpret or tune, with user-coded rules that can be combined and adjusted more directly. The proposed process is to start from a sample strategy, define factors and selection conditions, backtest individual rules, inspect daily holdings and returns, and revise the rules based on the strongest and weakest picks. It also recommends changing the test period to see whether results persist.

The article says its sample strategy held a concentrated position for two days and advertises strong historical returns, but the excerpt supplies no methodology, benchmark, transaction-cost treatment, or validation details for those claims. It explicitly warns that the instructions apply to an older platform version and are no longer suitable for the current platform. The workflow is introductory guidance, not evidence that the sample strategy will perform in live markets.

Key ideas

• The article contrasts automatically trained AI strategies with user-defined stock-selection rules.
• A custom strategy can combine factor conditions and can be tuned by changing its selection logic.
• The suggested workflow is to test individual rules, inspect holdings and returns, and refine the rules.
• Testing across different periods is proposed as a check on whether results persist.
• The article concerns an outdated platform version and does not provide enough detail to validate its advertised backtest results.

Tags

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