fs_identification_summary: Identification structure across a sweep

View source: R/fs_identification_summary.R

fs_identification_summaryR Documentation

Identification structure across a sweep

Description

Applies fs_identification_structure() across a named collection of bundles – typically one per trial size – and returns a single object the plot and table helpers consume. This is the package-level replacement for the inline fs_identification_figures*.qmd fragments.

Usage

fs_identification_summary(
  bundles,
  anchor,
  partner,
  proxy,
  confounders = NULL,
  sweep_label = "Trial size (n)",
  sg_def_col = "sg_def",
  covs_col = "covs"
)

Arguments

bundles

A named list. Each element is a results data frame, a bundle carrying results, or a path to an .rds. Names label the sweep points (for example "500" and "1000") and are used on the x axis in the order supplied.

anchor, partner, proxy

Character. The planted-structure triple.

confounders

Character vector. The analysis covariate pool.

sweep_label

Character. Axis label for the sweep dimension.

sg_def_col, covs_col

Character. Passed through.

Value

An object of class fs_identification: a list with per_point (the raw aggregates), tidy involvement and structure data frames, an accuracy data frame, and the triple it was built with.

Examples

## Not run: 
id <- fs_identification_summary(
  list(`500` = "results/fs_..._n500_combined_1_500.rds",
       `1000` = "results/fs_..._n1000_combined_1_500.rds"),
  anchor = "er", partner = "meno", proxy = "age",
  confounders = c("er", "age", "meno", "pgr", "nodes", "size", "grade"))
plot_fs_identification_involvement(id)
plot_fs_identification_structure(id)
fs_identification_table(id)

## End(Not run)


forestsearch documentation built on Nov. 15, 2026, 9:06 a.m.