rf_helper: Helper function(randomforest).

In postDoubleR: Post Double Selection with Double Machine Learning

Description

Helper function that dispatches to random forest (grf) for the post-double estimation.

Usage

rf_helper(X, Y, W, orthog.boost = FALSE, tree.n = 2000, tune = T,
  clustered = NULL)

Arguments

X	A matrix of covariates (must be all numeric)
Y	A vector of the target variable, of same length as the number of rows of Y, must be numeric
W	A vector of the treatment variable, of same length as the number of rows of X, must be numeric
orthog.boost	Whether to use orthogonal boosting, defaults to FALSE.
tree.n	Controls the number of trees grown, defaults to 2000.
tune	Whether to use hyperparameter tuning.
clustered	Whether to use cluster robust forests, defaults to FALSE.

Details

If you need to use something that is not a default argument here, please refer to custom_generator. Not using honesty is heavily advised, though, as that could lead to a very high splitting, and the honesty is used for essentialy the same reason as crossfitting.

Value

A list with two elements: The fitted W model and the fitted Y model.

Examples

  n = 2000; p = 10
  X = matrix(rnorm(n*p), n, p)
  W = rbinom(n, 1, 0.4 + 0.2 * (X[,1] > 0))
  Y = pmax(X[,1], 0) * W + X[,2] + pmin(X[,3], 0) + rnorm(n)



rf_helper( X = X, Y = Y, W = W, tree.n = 10)

postDoubleR documentation built on Oct. 7, 2019, 5:05 p.m.