estimate_NERM: Estimate nested error regression model (NERM)

In KatarzynaReluga/postcAIC: Post-cAIC selection inference under Linear Mixed Models

Description

Estimate NERM, store fixed and random parameters and variances

Usage

estimate_NERM(X, y, clusterID, X_cluster = NULL)

Arguments

X	Matrix with covariates for fixed effects
y	Vector of responses
clusterID	Vector with cluster labels
X_cluster	Matrix with cluster level covariates for fixed effects. Default: NULL.

Value

List with parameters:

fit_model_fixed	Estimated model using fixed effects
fit_model_mixed	Estimated model using fixed and random effects
sig_u	Variance parameter of random effects
sig_e	Variance parameter of errors
beta	Fixed effects (regression parameters)
u	Random effects
mu	Mixed effects

Examples

n = 10
m_i = 5
m_total = 50

clusterID = rep(1:5, n)
p = 10
beta = rep(2, p)
u_i = rnorm(n, 0, 2)
u_i_aug = rep(u_i, each = m_i)
X = matrix(rnorm(m_total * p), m_total, p)
y = X%*%beta + u_i_aug + rnorm(m_total, 0, 1)
fit_nerm <- estimate_NERM(X = X, y = y,
                          clusterID = clusterID,
                          X_cluster = NULL)

KatarzynaReluga/postcAIC documentation built on Jan. 25, 2022, 12:33 a.m.