gradient-forest

This repository is in an 'alpha' state, and is actively under development. We expect to make continual improvements to performance and usability.

Authors

This package is written and maintained by Julie Tibshirani ([email protected]), Susan Athey, and Stefan Wager.

The repository first started as a fork of the ranger repository -- we owe a great deal of thanks to the ranger authors for their useful and free package.

Installation

We recommend installing the package through devtools:

devtools::install_github("swager/gradient-forest", subdir = "r-package/gradient.forest")

Currently only Mac OS X and Linux are supported. Note that a compiler that implements C++11 is required (clang 3.3 or higher, or g++ 4.8 or higher).

Usage Examples

library(gradient.forest)

# Generate data.
n = 2000; p = 10
X = matrix(rnorm(n*p), n, p)
X.test = matrix(0, 101, p)
X.test[,1] = seq(-2, 2, length.out = 101)

# Perform quantile regression.
Y = X[,1] * rnorm(n)
q.forest = quantile.forest(X, Y, quantiles=c(0.1, 0.5, 0.9))
q.hat = predict(q.forest, X.test)
plot(X.test[,1], q.hat[,1], ylim = range(q.hat), xlab = "x", ylab = "y", type = "l")
lines(X.test[,1], q.hat[,2]); lines(X.test[,1], q.hat[,3])
abline(0, qnorm(0.9), lty = 2, col = 2); abline(0, -qnorm(0.9), lty = 2, col = 2)

# Perform treatment effect estimation.
W = rbinom(n, 1, 0.5)
Y = pmax(X[,1], 0) * W + X[,2] + pmin(X[,3], 0) + rnorm(n)
tau.forest = causal.forest(X, Y, W)
tau.hat = predict(tau.forest, X.test)
plot(X.test[,1], tau.hat$predictions, ylim = range(tau.hat$predictions, 0, 2), xlab = "x", ylab = "tau", type = "l")
lines(X.test[,1], pmax(0, X.test[,1]), col = 2, lty = 2)

# Add confidence intervals -- growing more trees is now recommended.
tau.forest = causal.forest(X, Y, W, num.trees = 4000)
tau.hat = predict(tau.forest, X.test, estimate.variance = TRUE)
sigma.hat = sqrt(tau.hat$variance.estimates)
plot(X.test[,1], tau.hat$predictions, ylim = range(tau.hat$predictions + 1.96 * sigma.hat, tau.hat$predictions - 1.96 * sigma.hat, 0, 2), xlab = "x", ylab = "tau", type = "l")
lines(X.test[,1], tau.hat$predictions + 1.96 * sigma.hat, col = 1, lty = 2)
lines(X.test[,1], tau.hat$predictions - 1.96 * sigma.hat, col = 1, lty = 2)
lines(X.test[,1], pmax(0, X.test[,1]), col = 2, lty = 1)

More usage examples, including examples around instrumental variables, can be found in the experiments directory.

References

Susan Athey, Julie Tibshirani and Stefan Wager. Solving Heterogeneous Estimating Equations with Gradient Forests, 2016. [arxiv]