common_mixture_rules: Estimate the union rule. This is the rule that covers all...

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

Estimate the union rule. This is the rule that covers all observations across the folds.

Description

This function takes in a list of rules that are grouped by variable sets. These rules for different variable sets may be slightly different across the folds so we make a union rule for each variable set. This entails creating a new rule that is essentially: (rule fold 1) OR (rule fold 2) OR (rule fold 3) etc. We then evaluate this rule on the input data to create a binary indicator of the union rule. Then, for variables used in the rule, find the min in and max in regions indicated by the rule for each variable. We then put these together with AND statements to create the union rule.

Usage

common_mixture_rules(
  group_list,
  data = data,
  mix_comps = mix_comps,
  mixture_results,
  n_folds
)

Arguments

group_list	List of dataframes grouped by rules for variable sets
data	Full data
mix_comps	Mixture components of A
mixture_results	data frame of results found for mixture rules
n_folds	Number of folds used in cross-validation

Value

Rules object. TODO: add more detail here.

Data frame with mixture results including the union rule and proportion found across the folds

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