compute_mse: Compute MSE

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

Description

Function compute_mse provides first-order correct MSE for mixed effects

Usage

compute_mse(C_cluster, X, sig_u, sig_e, clusterID, model = "NERM")

Arguments

C_cluster	Cluster-level covariates for fixed and random parameters
X	Matrix with covariates for fixed effects
sig_u	Variance parameter of random effects
sig_e	Variance parameter of errors
clusterID	Vector with cluster labels
model	Model we want to fit. Only NERM is supported for now.

Value

mse	MSE of mixed effects

Examples

n = 10
m_i = 5
m_total = 50

clusterID = rep(1:n, m_i)
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)

C_cluster = cbind(X[1:10, ], diag(n))

mse_first = compute_mse(C_cluster, X,
                        sig_u = fit_nerm$sig_u,
                        sig_e = fit_nerm$sig_e,
                        clusterID = clusterID)

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