Linear mixed-model estimation

Henderson's mixed-model equations, BLUE/BLUP via the V matrix, EM and REML estimation of the variance components.

Usage

morie_one_way(groups) 

morie_mme(X, Z, y, Sigma_inv, R_inv = NULL) 

morie_blue_blup_v(X, Z, y, Sigma, R = NULL) 

morie_em_lmm(X, Z, y, D0 = NULL, sigma2_0 = 1, n_iter = 200L, tol = 1e-10) 

morie_lmm_loglik(X, Z, y, D, R = NULL, beta = NULL) 

morie_reml_loglik(X, Z, y, D, R = NULL)

Arguments

y

numeric response vector

X

fixed-effect design matrix

Z

random-effect design matrix

groups

argument groups; not described in the source.

Sigma_inv

argument Sigma_inv; not described in the source.

R_inv

argument R_inv; not described in the source.

Sigma

argument Sigma; not described in the source.

R

argument R; not described in the source.

D0

argument D0; not described in the source.

sigma2_0

argument sigma2_0; not described in the source.

n_iter

argument n_iter; not described in the source.

tol

argument tol; not described in the source.

D

argument D; not described in the source.

beta

argument beta; not described in the source.

Value

A list holding the fitted quantities for the method called: point estimates, fitted values and any diagnostics it produces. See the individual method descriptions above for the components each returns.

Details

These functions mirror, in R, the implementations in the Python package morie; the two are checked against each other by the parity tests in tests/testthat/.

Examples

set.seed(1)
n <- 12; q <- 3
X <- cbind(1, rnorm(n))
Z <- diag(q)[sample(1:q, n, replace = TRUE), ]
u <- rnorm(q, 0, 1)
y <- as.numeric(X %*% c(1, 0.5) + Z %*% u + rnorm(n, 0, 0.3))
morie_mme(X, Z, y, Sigma_inv = diag(q))
#> $blue
#> [1] 1.7088626 0.5702941
#> 
#> $blup
#> [1] -0.15068478  0.11947382  0.03121096
#>