Package roadmap

Package roadmap

This page is a capabilities map for the R package: what the package can do, which exported entry points to use, and where to read next.

Website roadmap API reference

Core modeling capabilities

One-arm modeling (density + prediction)

• Bulk-only: dpmix() with a chosen bulk kernel and backend.
• Bulk + tail: dpmgpd() to splice a GPD tail beyond a threshold.
• Post-fit: summary(), params(), predict(), plot().

Read next: - Start: backends and workflow - Examples: one-arm templates

Causal (two-arm outcome modeling + optional PS)

• Fit: dpmix.causal() / dpmgpd.causal() (or lower-level builders when needed).
• Estimands: ate(), att(), qte(), qtt(), cate(), cqte(), and restricted-mean variants.
• Post-fit: predict() for arm and contrast quantities; plot() for effect diagnostics.

Read next: - Track: Causal - Theory: causal estimands and interpretation

Clustering (label-invariant summaries)

• Bulk-only clustering: dpmix.cluster()
• Tail-aware clustering: dpmgpd.cluster()
• Outputs: labels, assignment scores, and PSM-based diagnostics.

Read next: - Track: Clustering - Theory: clustering extension

Kernel and tail surface (what families are supported)

• Use kernel_support_table() for a compact “what works where” matrix.
• Use the kernels catalog for distributions and interpretation notes.

Read next: - Kernels hub - Track: Kernels & tails

Customization and extension (advanced)

• Customization lives primarily in param_specs (fixed/dist/link/link+dist modes).
• Extension is registry-based (kernel/tail registries), then flows through bundle → runner → consumers.

Read next: - Advanced: customization and tuning - Theory: customization maps + extension points - Developers hub

ImportantDeveloper notes

If you are extending the package (new kernels/tails, new predictors/plots), start from registry metadata and reuse shared bundle/runner layers instead of writing workflow-specific code.

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: Model Umbrella
• 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.