Skip to main content

Tools for statistical analysis of A/B test results

Project description

Tools for statistical analysis of A/B test results.

ABBA provides several statistical tools for analysis of binomial data, typically resulting from A/B tests:

  • Wald and Agresti-Coull confidence intervals on binomial proportions
  • Confidence intervals on the difference and ratio of two binomial proportions
  • Hypothesis tests for inequality of two binomial proportions
  • Multiple test correction for control of familywise error rate

Some simple example usage:

>>> import abba.stats
>>> abba.stats.confidence_interval_on_proportion(
...     num_successes=50, num_trials=200, confidence_level=0.99)
ValueWithInterval(value=0.25, lower_bound=0.17962262748069852, upper_bound=0.33643200973247306)

>>> experiment = abba.stats.Experiment(
...     num_trials=5, baseline_num_successes=50, baseline_num_trials=200)
>>> results = experiment.get_results(num_successes=70, num_trials=190)
>>> results.relative_improvement
ValueWithInterval(value=0.4736842105263157, lower_bound=-0.014130868125315277, upper_bound=0.90421878236700903)
>>> results.two_tailed_p_value

ABBA requires SciPy for underlying statistical functions.

For more info, see the docstrings, unit tests, and the ABBA website (including an interactive Javascript version) at

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Files for ABBA, version 0.1.0
Filename, size File type Python version Upload date Hashes
Filename, size ABBA-0.1.0.tar.gz (6.3 kB) File type Source Python version None Upload date Hashes View

Supported by

AWS AWS Cloud computing Datadog Datadog Monitoring Facebook / Instagram Facebook / Instagram PSF Sponsor Fastly Fastly CDN Google Google Object Storage and Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Salesforce Salesforce PSF Sponsor Sentry Sentry Error logging StatusPage StatusPage Status page