Make treatment decisions and compute estimated outcomes/treatment effects

Make treatment decisions and compute estimated outcomes/treatment effects

Usage

compute_estimates(
  df,
  Y_name,
  A_name,
  W_list,
  Z_list,
  k_fold_assign_and_CATE,
  nuisance_models,
  CATE_models,
  threshold,
  ps_trunc_level = 0.01,
  truncate_CATE
)

Arguments

df

dataframe containing 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 in nuisance models

Z_list

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

k_fold_assign_and_CATE

dataframe containing ids, fold assignments, and CATE estimate for each observation in df

nuisance_models

list of objects of class `Nuisance` containing outcome, treatment, and missingness SuperLearner models

CATE_models

list of discrete SuperLearner models for CATE from each fold

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.

ps_trunc_level

numeric level to use for truncation of any predicted values that fall below it

truncate_CATE

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

Value

List of objects of class `Results`. Each object contains the following for a given threshold:

aggregated_results

dataframe of overall results aggregated across `k` folds for given threshold

k_fold_results

list of results by fold for given threshold

decision_df

original dataset with decision made for each observation at a given threshold

k_non_na

folds that did not have proportion treated = 1 or = 0 (causing some effect estimates to be NA)