ci.adaptive_iptw {drtmle} | R Documentation |
Estimate confidence intervals for objects of class "adaptive_iptw"
## S3 method for class 'adaptive_iptw' ci(object, est = c("iptw_tmle"), level = 0.95, contrast = NULL, ...)
object |
An object of class |
est |
A vector indicating for which estimators to return a
confidence interval. Possible estimators include the TMLE IPTW
( |
level |
The nominal coverage probability of the desired confidence interval (should be between 0 and 1). Default computes 95% confidence intervals. |
contrast |
Specifies the parameter for which to return confidence
intervals. If |
... |
Other options (not currently used). |
An object of class "ci.adaptive_iptw"
with point estimates and
confidence intervals of the specified level.
# load super learner library(SuperLearner) # fit adaptive_iptw set.seed(123456) n <- 200 W <- data.frame(W1 = runif(n), W2 = rnorm(n)) A <- rbinom(n, 1, plogis(W$W1 - W$W2)) Y <- rbinom(n, 1, plogis(W$W1 * W$W2 * A)) fit1 <- adaptive_iptw( W = W, A = A, Y = Y, a_0 = c(1, 0), SL_g = c("SL.glm", "SL.mean", "SL.step"), SL_Qr = "SL.glm" ) # get confidence intervals for each mean ci_mean <- ci(fit1) # get confidence intervals for ATE ci_ATE <- ci(fit1, contrast = c(1, -1)) # get confidence intervals for risk ratio # by inputting own contrast function # this computes CI on log scale and back transforms myContrast <- list( f = function(eff) { log(eff) }, f_inv = function(eff) { exp(eff) }, h = function(est) { est[1] / est[2] }, fh_grad = function(est) { c(1 / est[1], -1 / est[2]) } ) ci_RR <- ci(fit1, contrast = myContrast)