CausalMixGPD

CausalMixGPD Documentation

Bayesian semiparametric modeling for heavy-tailed outcomes with Dirichlet-process mixtures and optional GPD tails.

Start Examples Kernels API Reference

Note

Use the Website roadmap for learning and the pkgdown reference for exact function interfaces.

Website roadmap Package roadmap Tracks

What this package is for

CausalMixGPD is a Bayesian semiparametric toolkit for outcome modeling when simple parametric families are not enough. It is designed for:

• Flexible bulk density via Dirichlet-process mixtures (multi-modality, skew, heavy shoulders)
• Extreme-tail behavior via an optional spliced Generalized Pareto tail
• Causal inference by fitting treated/control outcome models and reporting mean/quantile/conditional effects
• Clustering through label-invariant posterior summaries (labels + PSM diagnostics)

This site is workflow-first: it focuses on what to do next and how to interpret outputs, while pkgdown remains the source of truth for interfaces.

ImportantDeveloper notes

If you are extending the package, start from the Developers hub and the Registry pages; they describe the extension points and contracts that keep the rest of the site and S3 methods consistent.

Model / Workflows / Implementation

Model

Understand bulk vs tail architecture and kernel families.

Model architecture

Workflows

Follow ordered paths from first run to causal analysis.

Start roadmap

Implementation

Use developer pages and API docs for extension work.

Developers hub

Choose your path

Applied

Start from minimal run and move to example templates.

Start here Examples hub

Method

Focus on model choices, kernels, and advanced architecture.

Kernels hub Advanced hub

Developer

Work from architecture, tools, and contracts.

Developers hub API reference

Docs live in two places

• Quarto (docs/): curated navigation, workflows, kernel catalog, and architecture pages.
• pkgdown (docs/pkgdown/): primary function reference and package reference pages.

Quarto reference hub pkgdown reference

Small links: Cite | GitHub

Prereqs

• Required packages and data for this page are listed in the setup chunks above.

Outputs

• This page renders model fits, diagnostics, and summary artifacts generated by package APIs.

Interpretation

• Canonical concept page: Roadmap
• Treat this page as an application/example view and use the canonical page for core definitions.

Next

• Continue to the linked canonical concept page, then return for implementation-specific details.