# conditional_means: Estimate mean rewards mu for each treatment a In policytree: Policy Learning via Doubly Robust Empirical Welfare Maximization over Trees

## Description

μ_a = m(x) + (1-e_a(x))τ_a(x)

## Usage

 ``` 1 2 3 4 5 6 7 8 9 10``` ```## S3 method for class 'causal_forest' conditional_means(object, ...) ## S3 method for class 'instrumental_forest' conditional_means(object, ...) ## S3 method for class 'multi_arm_causal_forest' conditional_means(object, outcome = 1, ...) conditional_means(object, ...) ```

## Arguments

 `object` An appropriate causal forest type object `...` Additional arguments `outcome` Only used with multi arm causal forets. In the event the forest is trained with multiple outcomes Y, a column number/name specifying the outcome of interest. Default is 1.

## Value

A matrix of estimated mean rewards

## Methods (by class)

• `causal_forest`: Mean rewards μ for control/treated

• `instrumental_forest`: Mean rewards μ for control/treated

• `multi_arm_causal_forest`: Mean rewards μ for each treatment a

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18``` ```# Compute conditional means for a multi-arm causal forest n <- 500 p <- 10 X <- matrix(rnorm(n * p), n, p) W <- as.factor(sample(c("A", "B", "C"), n, replace = TRUE)) Y <- X[, 1] + X[, 2] * (W == "B") + X[, 3] * (W == "C") + runif(n) forest <- grf::multi_arm_causal_forest(X, Y, W) mu.hats <- conditional_means(forest) head(mu.hats) # Compute conditional means for a causal forest n <- 500 p <- 10 X <- matrix(rnorm(n * p), n, p) W <- rbinom(n, 1, 0.5) Y <- pmax(X[, 1], 0) * W + X[, 2] + pmin(X[, 3], 0) + rnorm(n) c.forest <- grf::causal_forest(X, Y, W) mu.hats <- conditional_means(c.forest) ```

policytree documentation built on July 7, 2021, 9:06 a.m.