Create an hte3 Task with a Production-Oriented Interface

hte_task() is the recommended high-level constructor for new users. It wraps make_hte3_Task_tx() with validated, production-oriented argument names.

Usage

hte_task(
  data,
  modifiers,
  confounders = modifiers,
  treatment,
  outcome,
  id = NULL,
  weights = NULL,
  treatment_type = c("default", "binomial", "categorical", "continuous"),
  propensity = NULL,
  outcome_regression = NULL,
  outcome_mean = NULL,
  propensity_learner = get_autoML(),
  outcome_learner = get_autoML(),
  mean_learner = NULL,
  cross_fit = TRUE,
  folds = 10,
  for_prediction = FALSE,
  ...
)

Arguments

data: A data frame or data.table containing treatment, outcome, modifiers, and confounders.
modifiers: Character vector of effect-modifier column names. These define the target modifier set V.
confounders: Character vector of adjustment-variable column names used for nuisance estimation. These define the adjustment set W.
treatment: Treatment column name.
outcome: Outcome column name.
id: Optional ID column name.
weights: Optional weight column name.
treatment_type: Treatment scale.
propensity: Optional matrix of user-supplied propensity estimates.
outcome_regression: Optional matrix of user-supplied outcome-regression estimates.
outcome_mean: Optional vector of user-supplied marginal outcome estimates.
propensity_learner: Learner used for propensity estimation when propensity is not supplied.
outcome_learner: Learner used for outcome-regression estimation when outcome_regression is not supplied.
mean_learner: Learner used for m(W) estimation when outcome_mean is not supplied.
cross_fit: Whether to cross-fit nuisance learners.
folds: Number of folds for nuisance estimation.
for_prediction: Whether to construct a prediction-only task with no nuisance fitting.
...: Additional arguments passed through to make_hte3_Task_tx().

Value

An hte3_Task object.

Details

When modifiers = V and confounders = W, the natural CATE target is E[Y(1) - Y(0) | V] = E[tau(W) | V]. In the supported binary/categorical-treatment setting, DR-, EP-, and the default two-stage T-learner align with that modifier-conditional target. The current R-learner does not generally target that object when V is a strict subset of W; it warns at fit time.

Examples

if (FALSE) { # \dontrun{
library(sl3)

data <- hte3_example_data(n = 80, seed = 1)

task <- hte_task(
  data = data,
  modifiers = c("W1", "W2"),
  confounders = c("W1", "W2", "W3"),
  treatment = "A",
  outcome = "Y",
  propensity_learner = Lrnr_mean$new(),
  outcome_learner = Lrnr_mean$new(),
  mean_learner = Lrnr_mean$new(),
  cross_fit = FALSE
)

task
} # }