Tidy an fetwfe fitted object

Returns a broom-style tidy data frame for an object of class "fetwfe". Row 1 is the overall ATT (term = "ATT"); subsequent rows are the cohort-specific ATTs (term = "Cohort <adoption-time>"), one per treated cohort, sorted by ascending cohort label. Standard error, z-statistic, and p-value reflect the value of se_type used at fit time (model-based by default, cluster-robust under se_type = "cluster"). Cohorts that the bridge penalty zeroed out (selected = FALSE) carry NA for std.error / statistic / p.value.

Usage

# S3 method for class 'fetwfe'
tidy(x, conf.int = TRUE, conf.level = 1 - x$alpha, ...)

Arguments

x

An object of class "fetwfe" returned by fetwfe().

conf.int

Logical. If TRUE (default), conf.low and conf.high columns are included.

conf.level

Numeric in (0, 1). Applies only to the overall-ATT row (row 1), whose CI is recomputed at this level; defaults to 1 - x$alpha (faithful to the alpha used at fit time; deviates from broom::tidy.lm's 0.95 default by design). The cohort rows pass through the fit-time catt_df bounds (reflecting the fit's ci_type and the fit-time alpha) and are NOT recomputed at conf.level (#197) — mirroring tidy.cohortStudy()'s stored-bounds behavior.

...

Unused; present for S3 compatibility.

Value

A data frame with G + 1 rows and columns term, estimate, std.error, statistic, p.value, optionally conf.low / conf.high, and selected (logical).

Details

The cohort-row conf.low / conf.high columns pass through the fit's stored catt_df bounds, so they reflect the fit's ci_type (#197): simultaneous (family-wise) by default, or pointwise when the fit used ci_type = "pointwise". They are NOT recomputed from conf.level (see the conf.level note). The overall-ATT row (row 1) is a scalar, so its CI is the pointwise Wald interval at conf.level (pointwise == simultaneous for a single effect).

Examples

if (FALSE) { # \dontrun{
  res <- fetwfeWithSimulatedData(
    simulateData(genCoefs(G = 3, T = 6, d = 2, density = 0.5, eff_size = 2),
                 N = 120, sig_eps_sq = 1, sig_eps_c_sq = 0.5, seed = 123)
  )
  broom::tidy(res)
} # }