Manski no-assumption bounds on the average treatment effect

Convenience form of morie_bndest taking the support as two scalars. Mirrors Python morie.fn.manski.

Usage

morie_manski(y, D, y_min, y_max)

Arguments

y: Outcomes.
D: Binary 0/1 treatment.
y_min, y_max: Assumed outcome support.

Value

The ATE list of morie_bndest.

References

Manski, C. F. (1990). Nonparametric bounds on treatment effects. American Economic Review, 80(2), 319-323.

Examples

set.seed(9)
y <- runif(60); D <- rbinom(60, 1, 0.5)
morie_manski(y, D, y_min = 0, y_max = 1)[c("ate_lower", "ate_upper")]
#> $ate_lower
#> [1] -0.4073182
#> 
#> $ate_upper
#> [1] 0.5926818
#>