Comprehensive Summary of Causal Forest Results

Produces a complete, publication-ready summary of a grf causal forest object, including: ATE with inference, heterogeneity assessment (BLP, GATES), CATE distribution, most important features, and policy-relevant summaries.

Usage

causal_forest_summary(
  cf,
  X,
  feature_names = NULL,
  n_groups = 5,
  top_features = 10,
  plot = TRUE
)

Arguments

cf

A causal_forest object from grf.

X

Data frame or matrix of covariates (same as used for fitting).

feature_names

Character vector. Names for the feature columns. If NULL, uses column names of X.

n_groups

Integer. Number of groups for GATES analysis. Default 5.

top_features

Integer. Number of top features to show in importance plot. Default 10.

plot

Logical. Produce plots. Default TRUE.

Value

A list with:

ate

Data frame: ATE estimate with SE, t-stat, p-value, CI.

cate_summary

Summary statistics of CATE distribution.

blp

BLP analysis results (from blp_analysis()).

feature_importance

Data frame of variable importance scores.

plot_cate_dist

ggplot2: CATE distribution histogram.

plot_importance

ggplot2: feature importance bar chart.

plot_blp

ggplot2: BLP coefficient plot.

plot_gates

ggplot2: GATES plot.

Examples

if (FALSE) { # \dontrun{
library(grf)
n <- 1000; p <- 10
X <- matrix(rnorm(n * p), n, p)
colnames(X) <- paste0("X", 1:p)
W <- rbinom(n, 1, 0.5)
tau <- pmax(X[, 1], 0)
Y   <- tau * W + rnorm(n)

cf  <- causal_forest(X, Y, W, num.trees = 500)
result <- causal_forest_summary(cf, X, feature_names = colnames(X))
result$ate
result$plot_cate_dist
} # }