sm.binomial.bootstrap:
Bootstrap goodness-of-fit test for a logistic regression model.
Description
This function is associated with sm.binomial
for the underlying fitting
procedure. It performs a Pseudo-Likelihood Ratio Test for the
goodness-of-fit of a standard parametric logistic regression of specified
degree
in the covariate x
.
Usage
sm.binomial.bootstrap(x, y, N = rep(1, length(x)), h, degree = 1, fixed.disp=FALSE, ...)
Arguments
x
vector of the covariate values
y
vector of the response values; they must be nonnegative integers.
h
the smoothing parameter; it must be positive.
N
a vector containing the binomial denominators.
If missing, it is assumed to contain all 1's.
degree
specifies the degree of the fitted polynomial in x
on the logit scale
(default=1).
fixed.disp
if TRUE
, the dispersion
parameter is kept at value 1 across the simulated samples, otherwise
the dispersion parameter estimated from the sample is used to generate
samples with that dispersion parameter (default=FALSE
).
Value
a list containing the observed value of the Pseudo-Likelihood Ratio Test
statistic, its observed p-value as estimated via the bootstrap method,
and the vector of estimated dispersion parameters when this value is not
forced to be 1.
Side Effects
Graphical output representing the bootstrap samples is produced on
the current graphical device.
The estimated dispersion parameter, the value of the test statistic
and the observed significance level are printed.Details
see Section 5.4 of the reference below.
References
Bowman, A.W. and Azzalini, A. (1997).
Applied Smoothing Techniques for Data Analysis:
the Kernel Approach with S-Plus Illustrations.
Oxford University Press, Oxford.