ortho_bounds_ss_wt: Compute orthogonal bounds (Semenova (2019))

In vsemenova/leebounds: Nonparametric bounds on Average Treatment Effect in the presence of Endogenous Selection

Compute orthogonal bounds (Semenova (2019))

Description

This function computes orthogonal bounds Semenova (2019) upper and lower bound on the Average Treatment Effect. Its input argument is a dataframe consisting of d=treat (binary treatment), s = selection (e.g., employment, test participation), outcome = sy observed only if s=1 (e.g., wage, test score). In addition, it also requires the first-stage estimates of the conditional probability of selection and the conditional quantile. It returns orthogonal (with respect to the first-stage) estimates of the upper and the lower bound. An optional vector of weights can be applied to the data to constructed weighted sample average.

Usage

ortho_bounds_ss_wt(
  leedata,
  treat_helps,
  y.hat,
  s.hat,
  weights = NULL,
  s_min = 0.1,
  ...
)

Arguments

leedata

data frame with treat, selection, outcome

Value

lower and upper bound on ATE

vsemenova/leebounds documentation built on Sept. 30, 2023, 8:30 a.m.