RJ2.gee: Corrected RJC for GEE

In wgeesel: Weighted Generalized Estimating Equations and Model Selection

Description

Calculate corrected RJC (Rotnitzky-Jewell information criterion) based on GEE with a modified robust variance estimator.

Usage

RJ2.gee(object)

Arguments

object

a fitted model object of class "wgee".

Value

Return the value of the corrected Rotnitzky-Jewell information criterion (RJC).

Note

RJ2.gee can only handle balanced data (data with dropout missingness). Two assumptions should be satisfied. (A1) The conditional variance of Y_{ij} given X_{ij} is correctly specified; (A2) A common correlation structure, Rc, exists across all subjects. If there is missingness, one can group the subjects by the cluter size of the response variable. And, calculate the modified robust variance in each group to get the pooled estimate of the variance.

Author(s)

Cong Xu, Zheng Li and Ming Wang

References

Rotnitzky, A. and Jewell, N.P., 1990. Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data. Biometrika, pp.485-497.

Wang, M. and Long, Q., 2011. Modified robust variance estimator for generalized estimating equations with improved small-sample performance. Statistics in Medicine, 30(11), pp. 1278-1291.

See Also

geeglm (geepack), RJ.gee

Examples

data(ohio)

fit <- wgee(resp ~ age + smoke+age:smoke, data=ohio, id=ohio$id, 
            family="binomial",corstr="exchangeable")
RJ2.gee(fit)       

wgeesel documentation built on May 2, 2019, 3:41 a.m.