c_hat_lower_RD: Lower Adaptive CI for the RD Parameter

In koohyun-kwon/rdadapt:

Lower Adaptive CI for the RD Parameter

Description

Calculates the lower end point of the adaptive CI for the RD parameter.

Usage

c_hat_lower_RD(
  delta,
  Cj,
  Cbar,
  Xt,
  Xc,
  mon_ind,
  sigma_t,
  sigma_c,
  Yt,
  Yc,
  tau,
  ht,
  hc
)

Arguments

delta	a nonegative scalar value: it can be left unspecified if ht and hc are specified.
Cj	the smoothness parameter aiming to adapt to.
Cbar	the largest smoothness parameter.
Xt	n_t by k design matrix for the treated units.
Xc	n_c by k design matrix for the control units.
mon_ind	index number for monotone variables.
sigma_t	standard deviation of the error term for the treated units (either length 1 or n_t).
sigma_c	standard deviation of the error term for the control units (either length 1 or n_c).
Yt	outcome value for the treated group observations.
Yc	outcome value for the control group observations.
tau	desired level of non-coverage probability.
ht	the modulus value for the treated observations; it can be left unspecified if delta is specified.
hc	the modulus value for the control observations; it can be left unspecified if delta is specified.

Details

This corresponds to the lower CI defined in Section 4.2 of our paper.

Value

the value of the lower end point of the adaptive CI.

Examples

n <- 500
d <- 2
X <- matrix(rnorm(n * d), nrow = n, ncol = d)
tind <- X[, 1] < 0 & X[, 2] < 0
Xt <- X[tind == 1, ,drop = FALSE]
Xc <- X[tind == 0, ,drop = FALSE]
mon_ind <- c(1, 2)
sigma <- rnorm(n)^2 + 1
sigma_t <- sigma[tind == 1]
sigma_c <- sigma[tind == 0]
Yt = 1 + rnorm(length(sigma_t), mean = 0, sd = sigma_t)
Yc = rnorm(length(sigma_c), mean = 0, sd = sigma_c)
c_hat_lower_RD(1, 1/2, 1, Xt, Xc, mon_ind, sigma_t, sigma_c,
Yt, Yc, 0.05)
c_hat_lower_RD(1, 1/2, Inf, Xt, Xc, mon_ind, sigma_t, sigma_c,
Yt, Yc, 0.05)

koohyun-kwon/rdadapt documentation built on May 8, 2022, 8:49 p.m.