mkVarCorr: Make Variance and Correlation Components of Covariance

In lme4: Linear Mixed-Effects Models using 'Eigen' and S4

Make Variance and Correlation Components of Covariance

Description

Computes from reCovs the variance and correlation components of random effect covariance matrices.

Usage

mkVarCorr(sc, cnms, nc = lengths(cnms, use.names = FALSE),
          theta, nms = names(cnms), reCovs = NULL, relReCovs = TRUE)

Arguments

sc	a numeric vector of length 1 giving the residual standard deviation for LMMs. Set to NULL for GLMMs.
cnms	a list of character vectors giving component dimnames, used only to determine nc and nms if those are missing.
nc	an integer vector giving component dimensions (number of columns).
theta	a numeric vector giving covariance parameters (lower triangular entries of relative Cholesky factors in column-major order), used only to determine reCovs when that is missing.
nms	a character vector giving component names.
reCovs	a list of Covariance objects.
relReCovs	a logical indicating if the components of reCovs represent relative covariance matrices (in which case they are scaled by sc^2).

Value

A list of matrices.

See Also

VarCorr.

lme4 documentation built on March 6, 2026, 1:07 a.m.