get_quant_g: Estimation of the quantiles of g(eta)

In matteobonvini/sensitivitypuc: Sensitivity Analysis via the Proportion of Unmeasured Confounding (PUC)

Description

get_quant_g returns the quantiles for the function g(eta)

Usage

get_quant_g(g, eps, min_g, max_g)

Arguments

g	a matrix n x delta g (generally evaluated at the X in the train set), where each column represents the function g(eta) for a given value of delta.
eps	vector of epsilon values (prop. unmeasured confounding)
min_g	the min value of g. The zero-quantile is set to be the min(g) - 1e-10, so that for eps = 0, the ATE is point-identified. This is because R takes the minimum to be the zero-quantile.
max_g	the max value of g. The one-quantile is set to be the max(g).

Value

a length(eps) x ncol(g) (= ndelta) matrix, where each entry (i,j) equals quantile(g[, delta[j]], p = eps[i])

See Also

if_gamma, get_ifvals, if_tau, get_g.

matteobonvini/sensitivitypuc documentation built on Dec. 9, 2020, 2:24 a.m.