Skip to contents

The primary human-readable output. When source is supplied, its worksheets, formatting, validation, and notes are preserved and a results sheet is added. path may be the same as source (the default fit workflow, which updates the supplied workbook) or a different path. Without source, a new workbook is constructed from inputs for directory inputs.

Usage

write_results_workbook(inputs, results, path, overwrite = FALSE, source = NULL)

Arguments

inputs

a list with obs, locs, and optionally target data frames, e.g. the result of read_inputs().

results

data.frame of summarized targets, as returned by summarize_targets().

path

character; destination .xlsx path.

overwrite

logical; TRUE to replace an existing file.

source

optional character path to an existing input workbook to amend.

Value

Invisibly, path.

Details

Each results row carries the target_id label and the resolved loc_id/cohort/age/dose identity, so a reader can trace it back to the target sheet row it came from. A single target-request row expands to one results row per location and per age in its span (see expand_targets()), which is why the estimates land on their own sheet rather than being appended to the request rows.

Examples

inputs <- list(
  obs = data.frame(
    obs_id = 1, loc_id = "A", cohort = 5, age = 5,
    dose = 2, positive = 3, sample_n = 10
  ),
  locs = data.frame(loc_id = "A", parent_id = NA)
)
res <- data.frame(
  target_id = "1", loc_id = "A", cohort = 5L, age = 5L, dose = 2L,
  n_draws = 100L, est_median = 0.8, est_lower = 0.7, est_upper = 0.9,
  ci_level = 0.95
)
out <- file.path(tempdir(), "results.xlsx")
write_results_workbook(inputs, res, out, overwrite = TRUE)