User guide

The user guide explains the full pySHT workflow: translate a scientific question into a test, prepare data that match the sampling design, call the procedure, and interpret the returned result. The same workflow applies across the package’s mean, variance, covariance, joint-parameter, distributional, goodness-of-fit, and structured-domain categories.

A practical route through the guide

1. Choose a test from the target quantity and sampling design.

2. Review data and assumptions, including independence, pairing, dimensionality, and distributional conditions.

3. Read test results to understand the R htest-style printed report, Monte Carlo uncertainty, log Bayes factors, and the fields available for programmatic use.

4. Follow reproducible inference when recording inputs, software versions, analysis options, and any random-number controls.

The API reference is organized by statistical category and contains the complete function signatures. The guide focuses on deciding what to call and what the result means.

• Choose a test
• Data and assumptions
• Test results
• Reproducible inference

Reference material

• Implemented methods

• Validation ledgers

• Migration from SHT for R

• Project information