Control Values for gls Fit

DESCRIPTION:

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 gls function.

USAGE:

glsControl(maxIter, msMaxIter, tolerance, msTol, msScale, msVerbose, 
           singular.ok, qrTol, returnObject, apVar, .relStep, 
           natural, natUnconstrained, sigma)  

OPTIONAL ARGUMENTS:

maxIter
maximum number of iterations for the gls optimization algorithm. Default is 50.
msMaxIter
maximum number of iterations for the ms optimization step inside the gls optimization. Default is 50.
tolerance
tolerance for the convergence criterion in the gls algorithm. Default is 1e-6.
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.
singular.ok
a logical value indicating whether non-estimable coefficients (resulting from linear dependencies among the columns of the regression matrix) should be allowed. Default is FALSE.
qrTol
a tolerance for detecting linear dependencies among the columns of the regression matrix in its QR decomposition. Default is .Machine$single.eps.
returnObject
a logical value indicating whether the fitted object should be returned when the maximum number of iterations is reached without convergence of the algorithm. Default is FALSE.
apVar
a logical value indicating whether the approximate covariance matrix of the variance-covariance parameters should be calculated. Default is TRUE.
.relStep
relative step for numerical derivatives calculations. Default is .Machine$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 ( corStruct and varFunc objects) used in the fit. If given as a list, it must have names 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 ( corStruct and varFunc objects) used in the fit. If given as a list, it must have names 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 residual standard error should be kept fixed during the optmization of the objective function. Defaults to NULL, in which case the residual 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 
glsControl(msMaxIter = 20, msVerbose = TRUE)