Propensity-score support overlap diagnostic (Cole-Hernan 2008)

Usage

mrm_check_overlap(data, treatment_col, covariates)

Arguments

data: data.frame.
treatment_col: Binary 0/1 treatment column.
covariates: Character vector of covariates.

Value

Named list with e_treated_quantiles, e_control_quantiles, common_support_lower, common_support_upper, n_outside_support, positivity_violations, interpretation.

Examples

set.seed(2026)
n <- 300L
x <- rnorm(n)
D <- rbinom(n, 1, plogis(0.5 * x))
df <- data.frame(D = D, age = x)
ovl <- mrm_check_overlap(df,
  treatment_col = "D",
  covariates = "age"
)
ovl$positivity_violations
#> [1] 0
ovl$interpretation
#> [1] "common support [0.229, 0.671]; 10 units outside; 0 positivity violations (e<.01 or e>.99)."