Bayes Factor Supplement

Use this page as a cautious guide to what a Bayes-factor supplement would mean conceptually, while keeping the current product support boundary explicit.

What this method is

This page is a concept guide for readers who want to think about Bayes factors as a supplement to conventional inferential summaries.

A Bayes-factor supplement is not the same thing as a replacement for every p-value-based result. In practice, it is usually discussed as an additional way to describe how strongly the observed data favor one model or hypothesis over another.

The current repo does not confirm a dedicated end-to-end Bayes-factor execution path, so this page should be read as orientation rather than as a product feature tutorial.

When to use or avoid

Use this page when:

  • you want to understand how a Bayes-factor style summary differs from a conventional significance summary
  • you are comparing alternative ways to express evidential strength
  • you need a conceptual placeholder for future specialized inference support

Avoid reading this page as evidence that Bayesian supplements are already available across the current product surface.

  • Do not assume this topic is exposed in every method page or export flow.
  • Do not treat this page as confirmation that the product currently computes Bayes factors for your analysis.

Required inputs

The current repo does not expose one confirmed public input contract for a Bayes-factor supplement.

Conceptually, a Bayes-factor workflow would usually need:

  • a clearly defined model or hypothesis comparison
  • data prepared for the target method family
  • assumptions about priors or other Bayesian setup choices

Those pieces are not yet documented in the current public product contract.

Outputs

No dedicated Bayes-factor output bundle was confirmed in this pass.

The safest public framing is:

  • this topic is part of the information architecture
  • the current implementation boundary is still incomplete
  • stronger output claims should wait for a confirmed route, schema, and test surface

No dedicated figure recommendation was confirmed for this topic in the current repo.

Interpretation notes

  • A Bayes factor is best understood here as a possible supplementary evidential summary, not as a product-wide default.
  • This page should not imply that Bayesian support is available everywhere.
  • Public wording should stay weaker than the page title might suggest until implementation evidence is stronger.

Limits

  • No dedicated Bayes-factor engine endpoint was confirmed in this pass.
  • No public result schema, prior configuration surface, or tested export-facing contract was confirmed.
  • This page is stronger as a conceptual guide than as a current implementation guide.

TODO (Phase02+)

  • Confirm implementation scope before making method-level support claims.