Skip to contents

Main function to calculate estimated treatment effects for treatment rule

Source: R/estimate_OTR.R
estimate_OTR.Rd

Primary function to estimate treatment effects for a given treatment rule. Can fit nuisance models internally or be provided with pre-fit nuisance models for the given dataset.

Usage

estimate_OTR(
  df,
  Y_name,
  A_name,
  W_list,
  Z_list,
  id_name = NULL,
  sl.library.CATE,
  nuisance_models = NULL,
  k_fold_assign_and_CATE = NULL,
  validRows = NULL,
  sl.library.outcome = NULL,
  sl.library.treatment = NULL,
  sl.library.missingness = NULL,
  threshold = c("0.05"),
  k_folds = 2,
  ps_trunc_level = 0.01,
  outcome_type = "gaussian",
  truncate_CATE = "TRUE"
)

Arguments

df

dataframe containing full dataset

Y_name

name of outcome variable in df

A_name

name of treatment variable in df

W_list

character vector containing names of covariates in the dataframe to be used for fitting nuisance models

Z_list

character vector containing names of variables in df used to fit CATE model (variables used in treatment rule)

id_name

name of participant ID variable

sl.library.CATE

character vector of SuperLearner libraries to use to fit the CATE model

nuisance_models

list of objects of class `Nuisance` containing outcome, treatment, and missingness SuperLearner models (only include if using pre-fit nuisance models)

k_fold_assign_and_CATE

dataframe containing ids, fold assignments, and CATE estimate for each observation in df (only include if using pre-fit nuisance models)

validRows

list containing validRows assignents for nuisance models using all observations (only include if using pre-fit nuisance models)

sl.library.outcome

character vector of SuperLearner libraries to use to fit the outcome models

sl.library.treatment

character vector of SuperLearner libraries to use to fit the treatment models

sl.library.missingness

character vector of SuperLearner libraries to use to fit the missingness models

threshold

character vector of decision thresholds for CATE to determine OTR. Values should be positive if `Y_name` is desirable outcome, negative if `Y_name` is undesirable outcome. If threshold is 0, use +0 for desirable, -0 for undesirable.

k_folds

integer number of folds to use for cross-validation (must specify if fitting outcome, treatment, and missingness models. Otherwise uses k from `k_fold_assign_and_CATE`)

ps_trunc_level

numeric evel below which propensity scores will be truncated (to avoid errors in computing AIPTW)

outcome_type

outcome_type specifying continuous (outcome_type = "gaussian") or binary (outcome_type = "binomial") outcome Y (if not providing pre-fit nuisance models)

truncate_CATE

logical to indicate if large CATE predictions should be truncated at -1 and 1 (default = TRUE)

Value

results

list of `Results` objects for each threshold. See description of `Results` object in `compute_estimates`

nuisance_models

list of `Nuisance` objects containing outcome, treatment, and missingness models used in each fold

CATE_models

CATE model used in each fold

Z_list

character vector containing names of variables in df used to fit CATE model (variables used in treatment rule)