.\" -*- nroff -*- generated from .Rd format
.BG
.FN lmeControl
.TL
Control Values for lme Fit
.DN
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.
.CS
lmeControl(maxIter, msMaxIter, tolerance, niterEM, msTol,
           msScale, msVerbose, returnObject, gradHess, apVar,
           .relStep, natural, natUnconstrained, sigma)
.OA
.AG maxIter
maximum number of iterations for the `lme'
optimization algorithm. Default is 50.
.AG msMaxIter
maximum number of iterations
for the `ms' optimization step inside the `lme'
optimization. Default is 50.
.AG tolerance
tolerance for the convergence criterion in the
`lme' algorithm. Default is 1e-6.
.AG niterEM
number of iterations for the EM algorithm used to refine
the initial estimates of the random effects variance-covariance
coefficients. Default is 25.
.AG 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. 
.AG msScale
scale function passed as the `scale' argument to
the `ms' function (see documentation on that function). Default
is `lmeScale'.
.AG msVerbose
a logical value passed as the `trace' argument to
`ms' (see documentation on that function). Default is
`FALSE'.
.AG 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'.
.AG 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'.
.AG apVar
a logical value indicating whether the approximate
covariance matrix of the variance-covariance parameters should be
calculated. Default is `TRUE'.
.AG .relStep
relative step for numerical derivatives
calculations. Default is `.Machine$double.eps^(1/3)'.
.AG 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'.
.AG 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'.
.AG 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').
.RT
a list with components for each of the possible arguments.

.SA
`lme', `ms', `lmeScale'
.EX
# 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)

.KW models
.WR
