Diagnosing Empty Backtest Data After Applying Stock Filters

Summary

This brief BigQuant support note explains a common cause of a “no data left after dropna” error when backtesting a strategy on the SSE 50 universe. The error can occur because the strategy’s filters leave no eligible stocks, rather than because the backtest engine has failed. The note points to a float-market-cap condition as an example: after applying a threshold below ten billion yuan, no stocks remain in the referenced case.

The practical diagnostic is to inspect the candidate universe after each screening condition and identify which filter empties it. The note offers links to example strategies but contains no code, detailed troubleshooting sequence, or backtest results. Its example is specific to the reported setup, so the threshold should not be treated as a general rule; other filters, dates, and data availability can also leave a universe empty.

Key ideas

• A backtest may report no data after missing-value removal because screening conditions leave no eligible securities.
• The note identifies a float-market-cap filter as the condition that emptied the referenced SSE 50 selection.
• Checking the remaining universe after each filter can help locate the cause.
• The example is setup-specific and does not establish a general market-cap threshold.

Tags

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