compute_CI: compute selective confidence intervals

In yiqunchen/PGInference: Powerful Graph Fused Lasso Inference

compute selective confidence intervals

Description

This function computes an equal-tailed (1-alpha) selective confidence intervals.

Usage

compute_CI(vTy, vTv, sigma, truncation, alpha = 0.05, steps_lim = 50)

Arguments

vTy	y: the observed response vector
vTv	v: the contrast vector that defines the parameter of interest
sigma	The known noise standard deviation. If unknown, we recommend a conservative estimate. If it is left blank, we use the sample variance as a conservative estimate.
truncation,	the truncation set (object: intervals) Computes a confidence interval for the mean of a truncated normal distribution.
alpha,	the significance level. Default to 0.05
steps_lim,	the maximum steps bisection method will take to initialize the LCB and UCB, default to 50.

Value

This function returns a vector of lower and upper confidence limits.

yiqunchen/PGInference documentation built on March 20, 2022, 11:51 p.m.