R/data.R

In RydenButler/attritR: Estimating average treatment effect under attrition on observables

#' Data with simulated attrition on unobservables.
#'
#' A dataset containing simulated data for computation with attritR.
#'
#' @format A data frame with 1000 rows and 4 variables:
#' \describe{
#'   \item{Y}{Response variable generated from the equation 
#'   Y = 1*Treatment + 1*Covariate + 0.25*Treatment*Covariate + U;
#'   where U is an unobserved error term. 
#'   Missingness in Y is generated by the equation R = -1*Treatment + 1*Covariate + 1*Instrument + V;
#'   where V is an unobserved error term that is correlated with U.
#'   Specifically, U and V are drawn randomly from a multivariate standard normal distribution with a covariance of 0.8.}
#'   \item{D}{Treatment assignment drawn randomly from a binomial distribution with equal probability.}
#'   \item{X}{Covariate generated from random draws on a uniform distribution from -1 to 1.}
#'   \item{Z}{Instrument generated from random draws on a uniform distribution from -1 to 2.}
#'   ...
#' }
"SimulatedAttrition"