Policy Learning Methods Statement — margot_policy_methods

Generates a boilerplate methods paragraph describing the policy learning workflow used for policy trees in `margot` (DR scores, tree fitting, honest evaluation, policy value contrasts with bootstrap CIs, treated-only uplift, and optional stability analysis).

Usage

margot_policy_methods_statement(
  object,
  depth = 2L,
  include_stability = TRUE,
  include_policy_value = TRUE,
  include_treated_only = FALSE,
  citations = TRUE,
  style = c("short", "long")
)

Arguments

object: A `margot_stability_policy_tree` (preferred) or a list with `$results` created by `margot_policy_tree()`.
depth: Integer; tree depth described in text (default 2).
include_stability: Logical; add details on the stability analysis if available (default TRUE).
include_policy_value: Logical; mention policy value contrasts and CIs (default TRUE).
include_treated_only: Logical; mention treated-only uplift and coverage (default FALSE).
citations: Logical; include inline citations (author-year) (default TRUE).
style: Character; one of "short" (1 paragraph) or "long" (multi-line) (default "short").

Value

A character scalar with the methods paragraph(s).