est_comb_exposure: Estimate the expected outcome for the combination of marginal...

In blind-contours/CVtreeMLE: Cross Validated Decision Trees with Targeted Maximum Likelihood Estimation

Estimate the expected outcome for the combination of marginal thresholds identified in the fold.

Description

Estimate the expected outcome given exposure to the combination of marginal exposures. This is different compared to the cumulative sum; whereas with the cumulative sum, the exposure is the additive effect of each marginal rule found in the fold, here each marginal rule is included in a Super Learner as a binary vector and therefore this can pick up possible nonlinearity between the combination of binary exposures.

Usage

est_comb_exposure(
  at,
  av,
  y,
  w,
  marg_rule_train,
  marg_rule_valid,
  no_marg_rules,
  aw_stack,
  family,
  parallel_cv,
  seed
)

Arguments

at	Training data
av	Validation data
y	Outcome variable
w	Vector of characters denoting covariates
marg_rule_train	Data frame of binary vectors for marginal rules identified in the training fold
marg_rule_valid	Data frame of binary vectors for marginal rules identified in the validation fold
no_marg_rules	TRUE/FALSE if no marginal rules were found across all
aw_stack	Super Learner library for fitting Q (outcome mechanism) and g (treatment mechanism)
family	Outcome type family
parallel_cv	TRUE/FALSE if parallel CV is used
seed	Seed number folds

Value

A list of the combination marginal results within a fold including:

• data: A data frame with the marginal rules evaluated as binary vectors, baseline covariates and predicted outcomes given ensemble fitting.

• learner: The Super Learner model fit to the data.

blind-contours/CVtreeMLE documentation built on June 22, 2024, 8:53 p.m.