ExpAn: A Python library for advanced statistical analysis ...
summary
ExpAn is an open source Python library (MIT, pip, GitHub) that creates an experiment object from metadata, features and KPIs and returns a pandas DataFrame result structure with sample sizes, means, uplift and confidence percentiles. It uses numpy and scipy, defaults to parametric tests, and offers a bootstrap option when assume_normal=False. Built analyses include delta tests, time-resolved trend analysis and subgroup breakdowns via a binning algorithm for subgroup discretization. The code scales to large web tests (examples show ~1,000–5,000 to ~1,000,000 entities) but does not handle assignment or concurrent-experiment interference yet.
AI-written summary · about
chapters (8)
- 0:00Introduction to ExpAn
- 4:33Library scope and design
- 6:40Experiment object structure
- 10:43Statistical analysis methods
- 15:25Result structure
- 17:24Summary and next steps
- 19:34Code demonstration
- 22:14Q&A session
related talks
28mExpAn - A Python library for advanced statistical analysis of A/B tests
01h 03mExperimentation in Data Science
30mStats Don't Have to Be Scary: Automatic a/b Test Analysis Using Python
01h 25mAdvanced Excel Analytics: Python Integration Workshop
01h 53mAnalyze your data in Python
22mExpEYES: Python Powered Open Source Pocket Science Lab
47mexpectation: A modern take on statistical A/B testing with e-values and martingales
01h 22mDoing frequentist statistics with Scipy