ExpAn - A Python library for advanced statistical analysis of A/B tests
summary
ExpAn is a Python library that defines a standard input format, standalone analysis functions and a generic result schema and ships on PyPI and GitHub; it measures uplift with bootstrap confidence intervals or parametric CIs, runs subgroup and time-resolved trend analyses, and implements a robust discretization for heavy-tailed KPIs. The library reports sample sizes, percent uplift and CI endpoints, records library version and warnings, and aims to add visualization; core routines are shown applied to multi-variant A/B experiments.
AI-written summary · about
chapters (14)
- 0:00ExpAn and A/B testing
- 2:31Speaker background
- 4:59A/B testing in business
- 6:26ExpAn for A/B testing
- 7:52ExpAn components
- 11:28Loading data in ExpAn
- 13:14Delta function for variants
- 15:24Trend analysis in experiments
- 17:38Subgroup analysis
- 18:40ExpAn results and metadata
- 22:36ExpAn open-source release
- 24:06Future visualization plans
- 25:07Closing remarks
- 25:48Q&A
related talks
32mExpAn: A Python library for advanced statistical analysis ...
30mStats Don't Have to Be Scary: Automatic a/b Test Analysis Using Python
01h 18mA/B Testing with SciPy
33mABBY: A Django app to document your A/B tests
26mAnti-Patterns in A/B Testing - or How does AB Testing Scale?
01h 25mAdvanced Excel Analytics: Python Integration Workshop
32mWrappers/Extenders- Companion Packages for Python Projects
22mExpEYES: Python Powered Open Source Pocket Science Lab