boot_grf

Inspired by the `boot` function in the bootstrap package with clusters + half-sampling added.
A future TODO could be to add parallel (not necessarily worth it)
https://stat.ethz.ch/R-manual/R-devel/library/parallel/doc/parallel.pdf

internal

Forest-based statistical estimation and inference.
GRF provides non-parametric methods for heterogeneous treatment effects estimation
(optionally using right-censored outcomes, multiple treatment arms or outcomes, or instrumental variables),
as well as least-squares regression, quantile regression, and survival regression,
all with support for missing covariates.

Erik Sverdrup

Generalized Random Forests

Julie Tibshirani

Susan Athey

Rina Friedberg

Vitor Hadad

David Hirshberg

Luke Miner

Stefan Wager

Marvin Wright

boot_grf function

<dl><dt>data</dt>
<dd>A data frame with the original data.</dd>
<dt>statistic</dt>
<dd>A function computing estimate(s) with signature (data, indices, ...) where
data is the original data, and indices a vector which defines the bootstrap sample.</dd>
<dt>R</dt>
<dd>The number of bootstrap replications.</dd>
<dt>clusters</dt>
<dd>Integer vector of cluster assignment, setting to 1:N corresponds to an ordinary
unclustered bootstrap.</dd>
<dt>half.sample</dt>
<dd>Whether to do half sample bootstrap (half the clusters are drawn). Default is TRUE.</dd>
<dt>...</dt>
<dd>Additional arguments passed on to statistic.</dd></dl>

Arguments

Simple clustered bootstrap. — boot_grf

<dl>

<dt>data</dt>
<dd>A data frame with the original data.</dd>


<dt>statistic</dt>
<dd>A function computing estimate(s) with signature (data, indices, ...) where
data is the original data, and indices a vector which defines the bootstrap sample.</dd>


<dt>R</dt>
<dd>The number of bootstrap replications.</dd>


<dt>clusters</dt>
<dd>Integer vector of cluster assignment, setting to 1:N corresponds to an ordinary
unclustered bootstrap.</dd>


<dt>half.sample</dt>
<dd>Whether to do half sample bootstrap (half the clusters are drawn). Default is TRUE.</dd>


<dt>...</dt>
<dd>Additional arguments passed on to statistic.</dd>

</dl>

Simple clustered bootstrap.

boot_grf: Simple clustered bootstrap.

Description

Usage

Value

Arguments

References