avar: Asymptotic variance
In koohyun-kwon/HTEBand: Uniform confidence band for heterogenous treatment effects

Description Usage Arguments Value Examples

Calculates the asymptotic variance expression given weights and conditional variance matrix.

1	avar(w.1, w.0, omega.1, omega.0, T.grad)

`w.1`	A `n_1` by `k` dimensional matrix of weight values corresponding to treated observations
`w.0`	A `n_0` by `k` dimensional matrix of weight values corresponding to control observations
`omega.1`	A `n_1` by `k` by `k` array of conditional variance corresponding to treated observations
`omega.0`	A `n_0` by `k` by `k` array of conditional variance corresponding to control observations
`T.grad`	A `k` dimensional gradient vector of `T_t f`

A scalar asymptotic variance value

avar(rep(1/50, 50), rep(1/50, 50), rep(1, 50), rep(1, 50), 1)
w.1 <- w.0 <- matrix(rep(1/50, 100), ncol = 2)
omega.1 <- omega.0 <- array(rep(c(1,0,0,1), each = 50), dim = c(50, 2, 2))
T.grad <- c(1, 1)
avar(w.1, w.0, omega.1, omega.0, T.grad)