SciencesPo (version 1.3.9)

stukel: Stukel's test of the logistic link

Description

The Stukel's test is an alternative to the goodness-of-fit test for logistic regression. It tests if significant change occurs in the model with the addition of new coefficients.

Usage

stukel(object, alternative = c("both", "alpha1", "alpha2"))

Arguments

object

An object of class glm.

alternative

add both z1 and z2 to model or just one of them.

encoding

UTF-8

Details

Two new covariates, z1 and z2 are generated such that $$z1 = 0.5 \* logit^{2} * I(pi >= 0.5)$$, $$z2 = - 0.5 \* logit^{2} \* I(pi <= 0.5)$$,="" where="" $$i(arg)="1$$" if="" arg="" is="" TRUE and $$I(arg) = 1$$ if FALSE.

References

Stukel, T.A. (1988) Generalized logistic models. Journal of the American Statistical Association 83: 426-431.

Hosmer, David W., et al (1997) A comparison of goodness-of-fit tests for the logistic regression model. Statistics in medicine 16.9, 965-980.

Allison, Paul (2014) Another Goodness-of-Fit Test for Logistic Regression.