R/causal_m_bias.R

#' M-Bias Data
#'
#' This dataset contains 100 observations, generated under the following mechanism:
#' U1 ~ N(0, 1)
#' U2 ~ N(0, 1)
#' Z ~ 8 U1 + U2 + N(0, 1) (measured factor)
#' X ~ U1 + N(0, 1) (exposure)
#' Y ~ X + U2 + N(0, 1) (outcome)
#'
#' @references D'Agostino McGowan L, Barrett M (2023). Causal inference is not a statistical problem. Preprint arXiv:2304.02683v1.
#'
#' @format A dataframe with 100 rows and 5 variables:
#'
#' * `u1`: an unknown factor
#' * `u2`: an unknown factor
#' * `covariate`: a known factor
#' * `exposure`: exposure
#' * `outcome`: outcome

"causal_m_bias"

Try the quartets package in your browser

Any scripts or data that you put into this service are public.

quartets documentation built on April 14, 2023, 12:25 a.m.