Generating Correct Performance Stats in Quantstats with Non-Daily Return Frequencies

[LubdhakM]

Summary
The current code implementation in quantstats appears to be tailored for daily return data. However, in lot of cases we see the trade log a stategy generate will have uneven frequencies causing quantstats to produce incorrect metrics when such data is used.

Request
Would it be feasible to update the module such a way which enables quantstats to handle various data uneven frequencies and accurately adjust all metrics accordingly? Or maybe a separate wrapper or module which takes in such uneven frequencies trade_log
along with underlying daily data and create a daily MTM.

Details
To illustrate, I’ve provided an example trade log below:

The trade log includes entry/exit timestamps, prices, and I also have access to 1-minute OHLC data for the instrument traded. My goal is to compute daily Mark-to-Market (MTM) values from this data, then use quantstats to create a performance tearsheet for the strategy.

Currently, it seems daily MTM data is necessary to generate an accurate Quantstats tearsheet. If there’s an alternative, more efficient approach to use this trade log for positional strategies in quantstats, suggestions would be highly appreciated.

[ranaroussi]

Closing this issue as it's older than 2025. The performance improvements and comprehensive compatibility layer in version 0.0.64 should address frequency-related performance stats issues. If you're still experiencing problems, please create a new issue with the latest version.