Diagnosing Missing Values in a BigQuant Simulation Data Pipeline

Summary

This forum post reports that a strategy based on a platform template runs, but simulation fails at a dropna step because of NaN values after the user adds factors. The author’s attempted fix was to fetch 90 calendar days of earlier daily index data, calculate returns from consecutive closes, and pass the date and return columns onward. The post includes a sample data pipeline and an experiment link, but contains no reply or confirmed resolution.

The example illustrates that extending the input history alone may not remove missing values. A shifted close calculation inherently leaves the first return undefined, while missing or mismatched dates, instruments, or factor inputs may also lead to nulls downstream. The document does not identify which cause triggered the reported failure, so it should be read as a troubleshooting question rather than a validated fix. It gives no backtest results or general solution for handling NaNs in simulation pipelines.

Key ideas

• The reported failure occurs during simulation at a dropna step after adding factors to a strategy template.
• The author tried loading 90 earlier calendar days of daily data before calculating returns.
• A return based on the previous close leaves an undefined value at the start of the series.
• The post provides an example pipeline but does not establish the cause or a working fix.
• Missing values should be traced through input dates, instruments, factor calculations, and downstream joins.

Tags

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