S_RF: honstRF Constructor

Description Arguments Format

View source: R/Shrf.R

Description

This is an implementation of the T-learner combined with honest random forest for both response functions

Arguments

feat

A data frame of all the features.

tr

A numeric vector contain 0 for control and 1 for treated variables.

yobs

A numeric vector containing the observed outcomes.

mtry

Number of variables to try at each node.

nodesize

...

replace

...

ntree

...

sample_fraction

...

nthread

...

splitratio

...

nodesizeAvg

...

alwaysTr

weather or not we always test weather we should split on the treatment assignment. Currently only alwaysTr=FALSE is implemented

Format

An object of class NULL of length 0.


soerenkuenzel/hte documentation built on June 12, 2018, 4:26 p.m.