R/simulated_data.R

In tbd: Estimation of Causal Effects with Outcomes Truncated by Death

#' @title Simulated data with known SACE
#'
#' @description This simulated dataset is to illustrate how to use \link[tbd]{sace} to estimate the SACE, and compare it with other naive methods. In this simulated data, by design, there is confounding between \code{Z} and \code{Y} caused by \code{X}, and confounding between \code{S} and \code{Y} caused by \code{X}.
#'
#' @docType data
#' @keywords datasets
#' @name simulated_data
#' @format A data frame with 5000 observations and 7 variables. \code{Z}, \code{A}, \code{Y}, \code{S} are 1-dimensional, and \code{X} is 3-dimensional. The variables are as follows:
#' \describe{
#'  \item{Z}{Binary treatment}
#'  \item{X.X1}{A factor covariate with 2 levels (1 and -1)}
#'  \item{X.V2}{A continuous covariate}
#'  \item{X.V3}{A contunuous covariate}
#'  \item{A}{The substitution variable which is continuous}
#'  \item{Y}{The continuous outcome. \code{NA} where \eqn{S = 0}}
#'  \item{S}{The survival indicator. \code{1} means survival and \code{0} means death.}
#' }
#' @source The dataset is generated by the simulation design of \cite{Wang et al. 2017} with \eqn{\delta_1 = 1} and \eqn{\delta_0 = 1}, which allows confounding between \code{Z} and \code{Y} caused by \code{X}, and confounding between \code{S} and \code{Y} caused by \code{X}.
#' @references Linbo Wang, Xiao-Hua Zhou, Thomas S. Richardson; Identification and estimation of causal effects with outcomes truncated by death, Biometrika, Volume 104, Issue 3, 1 September 2017, Pages 597-612, \url{https://doi.org/10.1093/biomet/asx034}

NULL