Fit a Heterogeneous Instrumental-Variable Forest — fit_instrumental

Wraps grf::instrumental_forest() with a column-name interface.

Usage

fit_instrumental_forest(
  data,
  outcome,
  treatment,
  instrument,
  covariates,
  num.trees = 2000,
  seed = 1,
  ...
)

Arguments

data: A data frame containing all required columns.
outcome: Name of outcome column.
treatment: Name of exposure or treatment column.
instrument: Name of instrument column.
covariates: Character vector of baseline covariate column names.
num.trees: Number of trees passed to grf::instrumental_forest().
seed: Random seed passed to grf::instrumental_forest().
...: Additional arguments passed to grf::instrumental_forest().

Value

A heteff_fit object.

Examples

if (FALSE) { # \dontrun{
set.seed(3)
n <- 400
z <- rbinom(n, 1, 0.5)
x1 <- rnorm(n)
x2 <- rnorm(n)
w <- 0.6 * z + 0.4 * x1 + rnorm(n)
y <- 1.2 * w + 0.5 * x2 + rnorm(n)
df <- data.frame(y = y, w = w, z = z, x1 = x1, x2 = x2)

fit <- fit_instrumental_forest(
  data = df,
  outcome = "y",
  treatment = "w",
  instrument = "z",
  covariates = c("x1", "x2")
)
head(as.data.frame(fit))
} # }