unsup_method_group_1: Use only subject 1's observations to construct prediction set...

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

Use only subject 1's observations to construct prediction set for a new subject 1 observation

Description

To check whether a potential y is in this interval, create an augmented sample containing subject 1's observations and y. As the nonconformity score for each observation in the augmented subject 1 data, compute the absolute difference between each observation in the augmented sample and the augmented mean. The p-value at y is the proportion of observations in the augmented sample with nonconformity score >= y's nonconformity score. The prediction set for a new observation on subject 1 is {y : p-value(y) >= alpha}.

Usage

unsup_method_group_1(Y, alpha, Y_new = NULL)

Arguments

Y	List containing data of all subjects. Each item in the list is a vector with one subject's observations. Only Y[[1]] is used.
alpha	Significance level
Y_new	New observation on subject 1

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.