Fit inverse-probability weights for a cohort data.frame, in memory (wrapper)

Frame-in / frame-out analogue of fit_trial_weights(): fits the IPW switching and/or IPCW censoring models in Rust from an in-memory cohort data.frame and returns the per-(id, period) factor table as a data.frame (id, period, weight_factor) — the table expand_trial_weighted_df() consumes. Wraps the extendr-generated fit_weights_df(). A bit64::integer64 id round-trips exactly (the returned id is integer64).

Usage

fit_trial_weights_df(
  cohort,
  id_col = "id",
  period_col = "period",
  treatment_col = "treatment",
  eligible_col = "eligible",
  outcome_col = "outcome",
  first_period = 0L,
  last_period = .Machine$integer.max,
  estimand = "PP",
  switch_numerator = NULL,
  switch_denominator = NULL,
  censor_col = NULL,
  censor_numerator = NULL,
  censor_denominator = NULL,
  pool_censor = "none"
)

Arguments

cohort

A data.frame (or tibble / data.table / Arrow Table) of long person-time rows. Coerced with as.data.frame().

id_col, period_col, treatment_col, eligible_col, outcome_col

Column names. Defaults match the TrialEmulation conventions.

first_period, last_period

Inclusive integer period bounds.

estimand

"ITT" or "PP".

switch_numerator, switch_denominator

Character vectors of covariate column names for the switching numerator / denominator models, or NULL to omit switching weights.

censor_col

Name of the {0,1} censoring-indicator column; the modelled response is 1 - censor_col. NULL omits IPCW weights.

censor_numerator, censor_denominator

Character vectors of covariate column names for the IPCW numerator / denominator models.

pool_censor

How the IPCW models are pooled across the previous-treatment strata: "none", "numerator", or "both".

Value

A data.frame with columns id, period, weight_factor.

Details

Model presence follows the same NULL-driven rule as fit_trial_weights(): a switching model is fitted when either switch_* covariate vector is non-NULL; an IPCW model is fitted when censor_col is non-NULL.

See also

fit_trial_weights() for the Parquet-path equivalent; expand_trial_weighted_fitted_df() to fit and expand in a single call.

Examples

# Fitting needs covariates, so load the shipped `data_censored` cohort:
if (requireNamespace("arrow", quietly = TRUE)) {
  cohort <- as.data.frame(arrow::read_parquet(system.file(
    "extdata", "weights", "input_data_censored.parquet", package = "tters")))
  factors <- fit_trial_weights_df(cohort, estimand = "PP",
    switch_numerator = "x2", switch_denominator = c("x2", "x1"))
}