ctseff: Estimating average effect curve for continuous treatment

In ehkennedy/npcausal: Nonparametric causal inference methods

Description

ctseff is used to estimate the mean outcomes in a population had all subjects received given levels of a continuous (unconfounded) treatment.

Usage

ctseff(y, a, x, bw.seq, sl.lib=c("SL.earth","SL.gam","SL.glm","SL.glmnet",
  "SL.glm.interaction","SL.mean","SL.ranger"))

Arguments

y	outcome of interest.
a	continuous treatment.
x	covariate matrix.
bw.seq	sequence of bandwidth values.
sl.lib	algorithm library for SuperLearner. Default library includes "earth", "gam", "glm", "glmnet", "glm.interaction", "mean", and "ranger".

Value

A list containing the following components:

res	estimates/SEs/CIs for population means.
bw.risk	estimated risk at sequence of bandwidth values.

References

Kennedy EH, Ma Z, McHugh MD, Small DS (2017). Nonparametric methods for doubly robust estimation of continuous treatment effects. Journal of the Royal Statistical Society, Series B. arxiv:1507.00747

Examples

n <- 500
x <- matrix(rnorm(n * 5), nrow = n)
a <- runif(n)
y <- a + rnorm(n, sd = .5)

ce.res <- ctseff(y, a, x, bw.seq = seq(.2, 2, length.out = 100))
plot.ctseff(ce.res)

# check that bandwidth choice is minimizer
plot(ce.res$bw.risk$bw, ce.res$bw.risk$risk)

ehkennedy/npcausal documentation built on Feb. 26, 2021, 2:43 a.m.