Tabulate scan predictions against planting date — planting_window

Reshapes a `scan_untested_envs()` result into one row per genotype x planting date, carrying the credible bounds and the envelope position. Useful on its own when you want the numbers rather than a plot.

Usage

planting_window_table(
  scan,
  dates,
  date_col = "date",
  genotypes = NULL,
  group = NULL
)

Arguments

scan: a `scan_untested_envs` object whose new "environments" are candidate planting dates.
dates: either (a) a data.frame with a column `env` matching the scan's new-environment names and a column giving the date, or (b) a vector of dates in the same order as the scan's new environments.
date_col: name of the date column when `dates` is a data.frame.
genotypes: optional character vector to subset genotypes.
group: optional named vector or data.frame (`gid`, `group`) assigning genotypes to groups, for group-mean curves.

Value

data.frame with `gid`, `env`, `date`, `yHat`, `lower`, `upper`, `ci_width`, `position`, `mahalanobis_ratio`, `prop_outside`, `weight_negativity`, and `group` when supplied.