find_common_marginal_rules: Create a new rule based on observations that meet every rule...

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

Create a new rule based on observations that meet every rule across the folds for each mixture.

Description

For each mixture component, a different rule could be found for each fold. Therefore, it is necessary to create one rule for each mixture component that can be interpreted as a common rule across the folds. To do this, observations that meet all rules for all folds are determined. Then a new rule is created for these observations. Specifically, we put OR statements between the rules found across the folds then look at the min and max values in this new region which encompasses all observations across the folds.

Usage

find_common_marginal_rules(
  fold_rules,
  data,
  mix_comps,
  marginal_results,
  n_folds
)

Arguments

fold_rules	List of rules found for each mixture component found across the folds
data	Full data which rules are evaluated
mix_comps	Vector of mixture components
marginal_results	Data frame holding the results for each marginal component rule
n_folds	Total number of folds

Value

Data frame with rules, threshold regions, proportion in folds and min/max values

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