cyberneticlibrary

Run A/B tests and hypothesis tests

statistical-analystpluginsetup L217,464
alirezarezvani/claude-skills
What it does

Run hypothesis tests, A/B analyses, and power calculations

Best for

Validating A/B experiment results with rigorous statistics instead of eyeballing conversion rates or running tests that are too small

Inputs
  • · sample data
  • · effect size
  • · confidence level
Outputs
  • · test result (p-value, confidence interval)
  • · power analysis
  • · sample size recommendation
Requires
  • · 3 stdlib Python tools (Z-test, t-test, chi-square, effect sizes, power analysis, Wilson score intervals)
Preconditions

Data collected from experiment or observational study; hypothesis formulated before test

Failure modes

Multiple-comparison bias (many tests, high false-positive rate); low power (experiment underpowered for real effect); wrong test chosen (parametric vs. non-parametric)

Trust signals
  • · 3 stdlib-only Python tools
  • · Covers Z-test, t-test, chi-square, effect sizes, power analysis, Wilson score intervals
  • · v2.9.0 stable release