multilevelMatching: Propensity Score Matching and Subclassification in...

In multilevelMatching: Propensity Score Matching and Subclassification in Observational Studies with Multi-Level Treatments

Description

multilevelMatching implements the estimators introduced in Yang et al. (2016) Propensity Score Matching and Subclassification in Observational studies with Multi-level Treatments: https://doi.org/10.1111/biom.12505. These are covariate- and propensity score-matching estimators for estimating the causal effect of multilevel treatment (i.e., 3 or more treatment types).

Details

The main function for estimation via matching on covariates or propensity scores is multiMatch. To carry out estimation via subclassification, use multilevelGPSStratification.

Examples

  sim_data <- multilevelMatching::simulated_data
  Y <- sim_data$outcome
  W <- sim_data$treatment
  X <- as.matrix(sim_data[ ,-(1:2)])
  names(Y) <- paste0("ID", 1:length(Y))

  trimming <- FALSE
  method <- c("covariates", "polr", "multinom")[2]

  multiMatch(Y,W,X,trimming=trimming,match_on=method)

multilevelMatching documentation built on May 8, 2019, 5:02 p.m.