Extract variance and correlation components
DESCRIPTION:
This function calculates the estimated variances, standard
deviations, and correlations between the random-effects terms in a
linear mixed-effects model, of class
lme
, or a nonlinear
mixed-effects model, of class
nlme
. The within-group error
variance and standard deviation are also calculated.
USAGE:
VarCorr(object, sigma, rdig)
REQUIRED ARGUMENTS:
- object
-
a fitted model object, usually an object inheriting from
class
lme
.
OPTIONAL ARGUMENTS:
- sigma
-
an optional numeric value used as a multiplier for the
standard deviations. Default is
1
.
- rdig
-
an optional integer value specifying the number of digits
used to represent correlation estimates. Default is
3
.
VALUE:
a matrix with the estimated variances, standard deviations, and
correlations for the random effects. The first two columns, named
Variance
and
StdDev
, give, respectively, the variance
and the standard deviations. If there are correlation components in
the random effects model, the third column, named
Corr
,
and the remaining unnamed columns give the estimated correlations
among random effects within the same level of grouping. The
within-group error variance and standard deviation are included as
the last row in the matrix.
EXAMPLES:
fm1 <- lme(distance ~ age, data = Orthodont, random = ~age)
VarCorr(fm1)