unsup_pool_cdfs: Unsupervised CDF pooling method

In RobinMDunn/ConformalTwoLayer: Distribution-Free Prediction Sets for Two-Layer Hierarchical Data

Unsupervised CDF pooling method

Description

Construct CDF pooling prediction interval for a new subject. At any value t and for each subject j, the empirical CDF is given by F.hat_j(t) = (1/n_j) sum_i=1^n_j I(Y_ji <= t). We let q.hat(alpha) = inf{t : (1/k) sum_j=1^k F.hat_j(t) >= alpha}. The CDF pooling prediction interval is [q.hat(alpha/2), q.hat(1-alpha/2)].

Usage

unsup_pool_cdfs(Y, alpha, new_Y = NULL)

Arguments

Y	List containing data of all subjects. Each item in the list is a vector with one subject's observations.
alpha	Significance level
new_Y	Observation on new subject

Value

List containing prediction interval size, prediction interval lower bound, prediction interval upper bound, and whether new observation is contained inside prediction interval.

RobinMDunn/ConformalTwoLayer documentation built on March 22, 2022, 6:38 p.m.