The values supplied in the function call replace the defaults and a
list with all possible arguments is returned. The returned list is
used as the
control argument to the
lme function.
maximum number of iterations for the
lme
optimization algorithm. Default is 50.
msMaxIter
maximum number of iterations
for the
ms optimization step inside the
lme
optimization. Default is 50.
tolerance
tolerance for the convergence criterion in the
lme algorithm. Default is 1e-6.
niterEM
number of iterations for the EM algorithm used to refine
the initial estimates of the random effects variance-covariance
coefficients. Default is 25.
msTol
tolerance for the convergence criterion in
ms,
passed as the
rel.tolerance argument to the function (see
documentation on
ms). Default is 1e-7.
msScale
scale function passed as the
scale argument to
the
ms function (see documentation on that function). Default
is
lmeScale.
msVerbose
a logical value passed as the
trace argument to
ms (see documentation on that function). Default is
FALSE.
returnObject
a logical value indicating whether the fitted
object should be returned when the maximum number of iterations is
reached without convergence of the algorithm. Default is
FALSE.
gradHess
a logical value indicating whether numerical gradient
vectors and Hessian matrices of the log-likelihood function should
be used in the
ms optimization. This option is only available
when the correlation structure (
corStruct) and the variance
function structure (
varFunc) have no "varying" parameters and
the
pdMat classes used in the random effects structure are
pdSymm (general positive-definite),
pdDiag (diagonal),
pdIdent (multiple of the identity), or
pdCompSymm (compound symmetry). Default is
TRUE.
apVar
a logical value indicating whether the approximate
covariance matrix of the variance-covariance parameters should be
calculated. Default is
TRUE.
.relStep
relative step for numerical derivatives
calculations. Default is
.Machine$double.eps^(1/3).
natural
a logical value, or a named list of logical values,
indicating whether a natural parameterization should be used for
the model structures, when the approximate covariance
matrix of the estimators is calculated. If given as a single logical
value, it is used for all model structures (
pdMat,
corStruct, and
varFunc objects) used in the fit. If
given as a list, it must have names
reStruct,
corStruct, and
varStruct corresponding to the model
structures used in the fit. Default is
TRUE.
natUnconstrained
a logical value, or a named list of logical
values, indicating whether an unconstrained parameterization should
be used for the natural parameters of the model structures. If
given as a single logical value, it is used for all model
structures (
pdMat,
corStruct, and
varFunc objects) used in the fit. If given as a list, it
must have names
reStruct,
corStruct, and
varStruct corresponding to the model structures used in the
fit. Default is
TRUE.
sigma
a numeric value indicating the value at which the
within-group standard error should be kept fixed during the
optmization of the objective function. Defaults to
NULL, in
which case the within-group standard error is estimated together
with the other model parameters. Must be a non-negative numeric
value - setting it to zero has the same effect as the default
(
NULL).
VALUE:
a list with components for each of the possible arguments.
SEE ALSO:
,
,
EXAMPLES:
# decrease the maximum number iterations in the ms call and
# request that information on the evolution of the ms iterations be printed
lmeControl(msMaxIter = 20, msVerbose = TRUE)