Methods for the class mmdml for generics from lme4.
fixef(object, ...)
# S3 method for mmdml
fixef(object, ...)ranef(object, ...)
# S3 method for mmdml
ranef(object, ...)
VarCorr(x, sigma = 1, ...)
# S3 method for mmdml
VarCorr(x, ...)
vcov(object, ...)
# S3 method for mmdml
vcov(object, ...)
An object of class mmdml. This object usually results
from a function call to mmdml.
See lmer from package lme4.
Further arguments passed to or from other methods.
See lmer from package lme4.
fixef.mmdml:
Extracts the estimator of the linear coefficient \(\beta_0\), which
is a named and numeric vector.
ranef.mmdml:
Extracts the random_eff entry from object.
VarCorr.mmdml:
The variance and correlation components are computed with the
sigma and the theta entries of x as in
lmer.
For each of the S repetitions, sigma and theta
computed
on the K sample splits are aggregated by taking the mean.
Then, the S mean-aggregated estimates are aggregated by
the median.
The variance and correlation components are computed with these
median-aggregated estimates.
vcov.mmdml:
It returns the variance-covariance matrix of the estimator of the linear
coefficient is extracted.
It is computed based on the asymptotic Gaussian distribution
of the estimator.
First, for each of the S repetitions, the variance-covariance
matrices computed
on the K sample splits are aggregated by taking the mean.
Second, the S mean-aggregated estimates are aggregated by
adding a term correcting for the randomness in the sample splits
and by taking the median of these corrected terms.
This final corrected and median-aggregated matrix is returned.
# NOT RUN {
## See example(mmdml) for examples
# }
Run the code above in your browser using DataLab