Classes of objects that result from fitting a parametric censored regression
model.
GENERATION:
This class of objects is returned by the
censorReg function
to represent a fitted parametric censored regression or survival model.
METHODS:
Objects of this class have methods for the functions
print,
summary
,
predict,
coefficients,
residuals,
plot and
anova.
STRUCTURE:
The following components must be included in a legitimate
censorReg object.
The residuals, fitted values, coefficients and effects should be extracted
by the generic functions of the same name, rather than
by the
"$" operator.
ARGUMENTS:
coefficients
the coefficients of the
linear.predictors, which multiply the
columns of the model matrix. It does not include the estimate of error
(sigma). The names of the coefficients are the names of the
single-degree-of-freedom effects (the columns of the model matrix).
distribution
a character string giving the distribution used in fitting the model.
family
a 2 element character vector giving the name of the family and
the link; used for internal purposes.
threshold
the value of the threshold parameter for the failure times. This will be
0 if not specified.
parms
the parameters of the model that are not coefficients of the model matrix.
The first of these will always be
log(sigma).
fixed
a logical vector the same length as
parms, where T indicates a parameter
that was fixed at its starting value and was not part of the iteration.
var
the final covariance matrix of the estimated coefficients including
regression coefficients and free parameters.
df.total
total degrees of freedom including adjustment for case weights.
dresiduals
the deviance residuals.
stzd.residuals
the standardized residuals.
deviance
minus twice the difference between the maximized log-likelihood under the
fitted model and a saturated model. Similar to the residual sum of squares.
null.deviance
the deviance corresponding to the model with only an intercept term, and
with
parms fixed at their final values.
loglik
the log-likelihood for the final and null models.
case.weights
the weights or frequencies used in fitting the model.
censor.codes
the censoring codes used in the fit.
n.censored
number of censored observations.
first.derivative
a vector of first derivatives of the likelihood with respect to the parameters
evaluated at the estimated parameters.
The object will also have the components of a
glm object not mentioned
above:
linear predictors,
fitted.values,
residuals,
effects,
R,
rank,
assign,
df.residual,
weights,
call,
contrasts,
terms
and
formula. See
glm.object.