VarCorr {nlme} | R Documentation |
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.
VarCorr(x, sigma, rdig)
x |
a fitted model object, usually an object inheriting from
class lme .
|
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 .
|
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.
Jose Pinheiro Jose.Pinheiro@pharma.novartis.com and Douglas Bates bates@stat.wisc.edu
fm1 <- lme(distance ~ age, data = Orthodont, random = ~age) VarCorr(fm1)