resampMakeFitObjResults(resampCall, fitCall, frame.eval, model.method=NULL) resampMakeFitObjResults.censorReg(resampCall, frame.eval)
fit
is the output a modeling function, like
lm
.
call
component or attribute of the
fit
argument in
resampCall
.
resampCall
and
fitCall
can be found. See
.
Typically the frame of the function that called the resample function.
"glm"
,
for example). In this
case it is assumed that the fitting routine has a
model.list
method (e.g.
),
and that the default method
for the fitting routine
accepts argument
method="model.list"
.
has no such method.
This function allows bootstrap and jackknife to recognize the syntax
#
bootstrap(fit, ...)
#
where
fit
is a fitted object (typically output of a modeling function).
This function
creates and then evaluates a new call with syntax appropriate to the
default method of the resample function, making use of
model.list
methods
for fitting routines if they are available. Here, a fitted object is
defined to be any object not of class
model.list
with component or
attribute
call
, and such that
call
contains a non-null
data
argument.
Special consideration is given when the fitted object is of class
by
resampMakeFitObjResults.censorReg
because of special issues that
arise in the stratified-sampling case, when the object is of class
censorRegList
.