ci_bound: Confidence interval for the maximum or minimum individual...

In li-xinran/RIQITE: Randomization Inference for Quantiles of Individual Treatment Effects

Confidence interval for the maximum or minimum individual effect

Description

Obtain confidence interval for the maximum or minimum individual effect

Usage

ci_bound(
  Z,
  Y,
  alternative = "greater",
  method.list = list(name = "DIM"),
  Z.perm = NULL,
  nperm = 10^3,
  impute = "control",
  alpha = 0.05,
  tol = 10^(-3)
)

Arguments

Z	An n dimensional treatment assignment vector.
Y	An n dimensional observed outcome vector.
alternative	A character takes value "greater" and "less" indicating the direction of the alternative hypothesis.
method.list	A list specifies the choice of the rank sum test statistic. For example, list(name = "DIM") means the difference-in-means statistic, list(name="Wilcoxon") means the Wilcoxon rank sum statistic, and list(name = "Stephenson", s = 10) means the Stephenson rank sum statistic with parameter s=10.
Z.perm	A n \times nperm matrix that specifies the permutated assignments for approximating the null distribution of the test statistic.
nperm	A positive integer representing the number of permutations for approximating the randomization distribution of the test statistic
impute	A character specifies the form of the test statistic. "control" means we use the imputed control potential outcomes in the test statistic, and "both" means we use the imputed observed outcome in the test statistic.
alpha	A numeric object specifies the level of the confidence interval. In particular, the confidence level is 1-\alpha.
tol	A numerical object specifying the precision of the obtained confidence intervals. For example, if tol = 10^(-3), then the confidence limits are precise up to 3 digits.

Value

The p-value for testing the specified null hypothesis of interest.

li-xinran/RIQITE documentation built on July 1, 2023, 6:58 p.m.