heterogeneity_CLAN: Evaluate treatment effect heterogeneity along CLAN variables

In GenericML: Generic Machine Learning Inference

Evaluate treatment effect heterogeneity along CLAN variables

Description

This function tests for statistical significance of all CLAN difference parameters that were specified in the function setup_diff(). It reports all CLAN variables along which there are significant difference parameters, which corresponds to evidence for treatment effect heterogeneity along this variable, at the specified significance level.

Usage

heterogeneity_CLAN(x, learner = "best", significance_level = 0.05)

Arguments

x	An object of class "GenericML", as returned by the function GenericML().
learner	A character string of the learner whose CLAN generic target estimates are of interest. Default is "best" for the best learner for CLAN.
significance_level	Level for the significance tests. Default is 0.05.

Value

An object of class "heterogeneity_CLAN", consisting of the following components:

p_values

A matrix of p values of all CLAN difference parameters for all CLAN variables.

significant

The names of variables with at least one significant CLAN difference parameter ("variables"), their number "num_variables", and the total number of significant CLAN difference parameters "num_params". All significance tests were performed at level significance_level.

min_pval

Information on the smallest p value: Its value ("value"), the variable in which it was estimated ("variable"), the CLAN difference parameter it belongs to ("parameter"), and whether or not it is significant at level significance_level ("significant").

"learner"

Name of the learner whose median estimates we used for the listed results.

"significance_level"

The level of the significance tests.

See Also

GenericML()

GenericML documentation built on June 18, 2022, 9:09 a.m.