_pkgdown.yml

url: https://choxos.github.io/mlumr/

template:
  bootstrap: 5
  light-switch: true          # dark / light mode toggle (bslib)
  math-rendering: katex        # fast, crisp LaTeX in the math-heavy vignettes
  bslib:
    primary: "#1A6D73"         # mlumr brand teal (from the hex logo)
    link-color: "#1A6D73"
    base_font: {google: "Inter"}
    heading_font: {google: "Inter"}
    code_font: {google: "JetBrains Mono"}

navbar:
  structure:
    left: [intro, app, lesson, reference, articles, news]
    right: [search, lightswitch, github]
  components:
    # In-browser playground (separate deploy on the webapp branch at /mlumr/app),
    # featured prominently in the navbar.
    app:
      text: Web app
      icon: fa-rocket
      href: https://choxos.github.io/mlumr/app
      aria-label: mlumr web app (in-browser playground)
    # Narrated interactive lesson (separate deploy on the lesson branch at
    # /mlumr/lesson).
    lesson:
      text: Lesson
      icon: fa-graduation-cap
      href: https://choxos.github.io/mlumr/lesson
      aria-label: mlumr narrated interactive lesson
    # Articles dropdown with short outcome-type labels (the full descriptive
    # titles stay on the pages themselves and in the CRAN vignette index).
    articles:
      text: Articles
      menu:
      - text: Getting started
      - text: Introduction to mlumr
        href: articles/introduction.html
      - text: Preparing data and integration points
        href: articles/data-preparation.html
      - text: Outcome types
      - text: Binary outcomes
        href: articles/binary-outcomes.html
      - text: Continuous outcomes
        href: articles/continuous-outcomes.html
      - text: Count outcomes
        href: articles/count-outcomes.html
      - text: Survival (time-to-event) outcomes
        href: articles/survival-outcomes.html
      - text: Methods
      - text: Fitting, priors, and diagnostics
        href: articles/fitting-and-diagnostics.html
      - text: "Choosing a method for single-arm indirect comparisons"
        href: articles/choosing-a-method.html
      - text: Aggregate subgroup information for the relaxed model
        href: articles/subgroup-identification.html

articles:
- title: Getting started
  navbar: Getting started
  contents:
  - introduction
  - "data-preparation"
- title: Outcome types
  navbar: Outcome types
  contents:
  - "binary-outcomes"
  - "continuous-outcomes"
  - "count-outcomes"
  - "survival-outcomes"
- title: Methods
  navbar: Methods
  contents:
  - "fitting-and-diagnostics"
  - "choosing-a-method"
  - "subgroup-identification"

reference:
- title: Data setup
  desc: >
    Assemble the index IPD and comparator AgD into one `mlumr_data` object and
    build the quasi-Monte Carlo integration points for population adjustment.
  contents:
  - set_ipd
  - set_agd
  - set_agd_surv
  - combine_data
  - add_integration
  - check_integration
  - check_identification
  - unnest_integration
  - make_knots
- title: Covariate distributions
  desc: Marginal-distribution helpers for the comparator integration points.
  contents:
  - distr
  - qbern
  - pbern
  - dbern
  - GammaDist
  - logitNormal
- title: Model fitting
  desc: Fit the Bayesian ML-UMR model (SPFA / relaxed) and select the Stan backend.
  contents:
  - mlumr
  - mlumr_engine
- title: Priors
  desc: Prior constructors, the prior summary, and prior-sensitivity analysis.
  contents:
  - prior_normal
  - prior_student_t
  - prior_cauchy
  - prior_exponential
  - prior_summary
  - prior_sensitivity
  - default_prior_intercept
  - default_prior_beta
  - default_prior_sigma
  - default_prior_aux
  - default_prior_smooth
- title: Treatment effects and predictions
  desc: Population-standardized effects and absolute predictions in either population.
  contents:
  - marginal_effects
  - conditional_effects
  - predict.mlumr_fit
  - conditional_predict
- title: Frequentist benchmarks
  desc: Unadjusted (naive) and simulated-treatment-comparison (STC) reference estimators.
  contents:
  - naive
  - stc
- title: Model comparison and diagnostics
  desc: Compare SPFA against relaxed, and the information criteria behind it.
  contents:
  - check_diagnostics
  - compare_models
  - calculate_dic
  - calculate_loo
  - calculate_waic
- title: Plots
  desc: >
    The `plot()` methods, the Kaplan-Meier overlay, the method-comparison forest,
    and the prior-versus-posterior overlay.
  contents:
  - plot.mlumr_prediction
  - plot.mlumr_marginal_effects
  - plot.mlumr_conditional_effects
  - geom_km
  - mlumr_forest
  - plot_prior_posterior
- title: Example datasets
  desc: >
    Bundled index IPD and comparator AgD for the worked examples, one pair per
    outcome family (binary, continuous, count, survival).
  contents:
  - psoriasis_ipd
  - psoriasis_agd
  - shoulder_ipd
  - shoulder_agd
  - caries_ipd
  - caries_agd
  - ndmm_ipd
  - ndmm_agd
  - ndmm_agd_covs