Deriving Daily Returns from Cumulative Backtest Returns

Summary

This short BigQuant support exchange answers how to obtain an algorithm’s daily returns from backtest performance data. Its method reads the raw performance table, takes the cumulative algorithm period return, converts it to a growth index by adding one, and calculates the percentage change between successive observations. It fills the first missing change with zero, yielding a daily return series suitable for further analysis or export.

The post provides a transformation rather than a complete CSV-writing workflow: it does not show a file export step, define the frequency or indexing of the performance table, or discuss handling missing dates and irregular observations. The approach relies on the cumulative return field being correctly ordered and representing compounded performance. No worked example, validation, or comparison with another return calculation is included.

Key ideas

• Read cumulative algorithm returns from the raw backtest performance data.
• Convert cumulative returns into period returns by measuring percentage changes in one plus the cumulative return.
• The first missing change is filled with zero.
• The exchange does not show the final step for writing the resulting series to a CSV file.

Tags

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