Fit propensity scores for treatment contrasts

fit_treat_mech(
  train_data,
  valid_data = NULL,
  contrast,
  learners,
  m_names,
  w_names,
  type = c("g", "h"),
  bounds = c(0.01, 0.99)
)

Arguments

train_data: A data.table containing the observed data; columns are in the order specified by the NPSEM (Y, M, R, Z, A, W), with column names set appropriately based on the data. Such a structure is merely a convenience utility to passing data around to the various core estimation routines and is automatically generated medoutcon.
valid_data: A holdout data set, with columns exactly matching those appearing in the preceding argument train_data, to be used for estimation via cross-fitting. Optional, defaulting to NULL.
contrast: A numeric double indicating the two values of the intervention A to be compared. The default value of c(0, 1) assumes a binary intervention node A.
learners: Stack, or other learner class (inheriting from Lrnr_base), containing a set of learners from sl3, to be used in fitting a propensity score models, i.e., g := P(A = 1 | W) and h := P(A = 1 | M, W).
m_names: A character vector of the names of the columns that correspond to mediators (M). The input for this argument is automatically generated by medoutcon.
w_names: A character vector of the names of the columns that correspond to baseline covariates (W). The input for this argument is automatically generated by medoutcon.
type: A character indicating which of the treatment mechanism variants to estimate. Option "g" is the propensity score g(A|W) while option "h" is a re-parameterized mediator density h(A|M,W).
bounds: A numeric vector containing two values, the first being the minimum allowable estimated propensity score value and the second being the maximum allowable for estimated propensity score value.