Learn R Programming

grf (version 2.5.0)

Generalized Random Forests

Description

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.

Copy Link

Version

Install

install.packages('grf')

Monthly Downloads

16,665

Version

2.5.0

License

GPL-3

Issues

62

Pull Requests

0

Stars

1,047

Forks

266

Maintainer

Erik Sverdrup

Last Published

October 9th, 2025

Functions in grf (2.5.0)

get_scores.causal_survival_forest

Compute doubly robust scores for a causal survival forest.
get_scores.instrumental_forest

Doubly robust scores for estimating the average conditional local average treatment effect.
get_scores

Compute doubly robust scores for a GRF forest object
instrumental_forest

Intrumental forest
leaf_stats.instrumental_forest

Calculate summary stats given a set of samples for instrumental forests.
get_scores.causal_forest

Compute doubly robust scores for a causal forest.
ll_regression_forest

Local linear forest
leaf_stats.regression_forest

Calculate summary stats given a set of samples for regression forests.
leaf_stats.default

A default leaf_stats for forests classes without a leaf_stats method that always returns NULL.
grf_options

grf package options
grf-package

grf: Generalized Random Forests
merge_forests

Merges a list of forests that were grown using the same data into one large forest.
lm_forest

LM Forest
plot.grf_tree

Plot a GRF tree object.
leaf_stats.causal_forest

Calculate summary stats given a set of samples for causal forests.
plot.rank_average_treatment_effect

Plot the Targeting Operator Characteristic curve.
predict.multi_arm_causal_forest

Predict with a multi arm causal forest
predict.causal_forest

Predict with a causal forest
predict.boosted_regression_forest

Predict with a boosted regression forest.
predict.causal_survival_forest

Predict with a causal survival forest forest
print.boosted_regression_forest

Print a boosted regression forest
print.grf

Print a GRF forest object.
predict.instrumental_forest

Predict with an instrumental forest
multi_arm_causal_forest

Multi-arm/multi-outcome causal forest
print.rank_average_treatment_effect

Print the Rank-Weighted Average Treatment Effect (RATE).
multi_regression_forest

Multi-task regression forest
print.grf_tree

Print a GRF tree object.
predict.probability_forest

Predict with a probability forest
predict.lm_forest

Predict with a lm forest
predict.ll_regression_forest

Predict with a local linear forest
predict.multi_regression_forest

Predict with a multi regression forest
print.tuning_output

Print tuning output. Displays average error for q-quantiles of tuned parameters.
split_frequencies

Calculate which features the forest split on at each depth.
predict.survival_forest

Predict with a survival forest
rank_average_treatment_effect.fit

Fitter function for Rank-Weighted Average Treatment Effect (RATE).
predict.regression_forest

Predict with a regression forest
probability_forest

Probability forest
variable_importance

Calculate a simple measure of 'importance' for each feature.
regression_forest

Regression forest
survival_forest

Survival forest
tune_forest

Tune a forest
test_calibration

Omnibus evaluation of the quality of the random forest estimates via calibration.
tune_ll_regression_forest

Local linear forest tuning
predict.quantile_forest

Predict with a quantile forest
quantile_forest

Quantile forest
rank_average_treatment_effect

Estimate a Rank-Weighted Average Treatment Effect (RATE).
tune_ll_causal_forest

Local linear forest tuning
estimate_rate

Compute rate estimates, a function to be passed on to bootstrap routine.
average_treatment_effect

Get doubly robust estimates of average treatment effects.
causal_survival_forest

Causal survival forest
boosted_regression_forest

Boosted regression forest
causal_forest

Causal forest
best_linear_projection

Estimate the best linear projection of a conditional average treatment effect.
create_dot_body

Writes each node information If it is a leaf node: show it in different color, show number of samples, show leaf id If it is a non-leaf node: show its splitting variable and splitting value If trained with missing values, the edge arrow is filled according to which direction the NAs are sent.
get_tree

Retrieve a single tree from a trained forest object.
get_leaf_node

Find the leaf node for a test sample.
get_forest_weights

Given a trained forest and test data, compute the kernel weights for each test point.
generate_causal_data

Generate causal forest data
get_scores.multi_arm_causal_forest

Compute doubly robust scores for a multi arm causal forest.
generate_causal_survival_data

Simulate causal survival data
expected_survival

Compute E[T | X]
export_graphviz

Export a tree in DOT format. This function generates a GraphViz representation of the tree, which is then written into `dot_string`.
boot_grf

Simple clustered bootstrap.