unsup_generate_data: Generate data for unsupervised normal examples

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

Generate data for unsupervised normal examples

Description

Draw distribution-specific parameters theta_1, ..., theta_k ~ N(mu, tau^2). For each distribution j, draw observations Y_j1, ..., Y_jn_j ~ N(theta_j, sigma^2).

Usage

unsup_generate_data(k, n_vec, mu, tau_sq, sigma_sq = 1)

Arguments

k	Number of subjects
n_vec	Vector of length k, representing number of observations from each subject. We allow this to vary across subjects.
mu	Mean of parameter distribution
tau_sq	Variance of parameter distribution
sigma_sq	Variance of observations in given distribution

Value

A list of length k, where each item contains a vector of observations from a single distribution

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