pybaycor

Bayesian inference of (robust) correlation coefficients for uncertain and censored data, built on PyMC and ArviZ.

Contents

• Installation
  • With pip
  • With pixi (recommended for development)
  • From source with pip
• Usage
  • One class, one model
  • Fitting
  • Summaries
  • Plots
  • Censoring conventions
  • Standardisation and priors
• Dust scaling relations in nearby galaxies with pybaycor
  • The data
  • Step 1 - Two variables without uncertainties
  • Step 2 - Adding measurement uncertainties
  • Step 3 - Robustness against outliers
  • Step 4 - Upper limits (censored data)
  • Step 5 - The full problem: five variables, upper limits and missing data
  • Step 6 - Is a correlation significant?
  • Step 7 - Going further: partial correlations
  • Summary
  • References
• API Reference
  • pybaycor
• pybaycor modernisation plan
  • 1. Where the package stood (v0.2.0)
  • 2. Target architecture
  • 3. Statistical model
  • 4. Inference
  • 5. Open-issue map
  • 6. Remaining work (not on this branch)
• Changelog
  • [0.3.0] - 2026-10-21
  • [0.2.0] - 2021-03

Indices

• Index

• Module Index