sup_generate_data: Generate data for supervised normal examples

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

Generate data for supervised normal examples

Description

Draw distribution-specific parameters theta_1, ..., theta_k ~ N(mu, tau^2). For each distribution j, draw covariates X_j1, ..., X_jn_j ~ N(0, 1), and draw outcomes Y_ji = theta_j*X_ji + epsilon_ji, epsilon_ji ~ N(0, sigma^2).

Usage

sup_generate_data(k, n, mu, tau_sq, sigma_sq)

Arguments

k	Number of subjects
n	Number of observations from each subject. For our examples, we assume this is equal across subjects.
mu	Mean of parameter distribution
tau_sq	Variance of parameter distribution
sigma_sq	Variance of epsilon in response model

Value

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

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