conLin
an optional condensed linear model object, consisting of
a list with components "Xy"
, corresponding to a regression
matrix (X
) combined with a response vector (y
), and
"logLik"
, corresponding to the log-likelihood of the
underlying lme model. Defaults to attr(object, "conLin")
.
control
an optional list with control parameters for the
initialization and optimization algorithms used in
lme
. Defaults to list(niterEM=20, gradHess=TRUE)
,
implying that 20 EM iterations are to be used in the derivation of
initial estimates for the coefficients of the reStruct
component of object
and, if possible, numerical gradient
vectors and Hessian matrices for the log-likelihood function are to
be used in the optimization algorithm.