RJC: Rotnitzky-Jewell Information Criterion

In mcglm: Multivariate Covariance Generalized Linear Models

Rotnitzky–Jewell Information Criterion

Description

Computes the Rotnitzky–Jewell information criterion (RJC) for objects of class mcglm. This criterion is based on quasi-likelihood theory and is intended for model assessment in marginal models.

Usage

RJC(object, id, verbose = TRUE)

Arguments

object	An object of class mcglm representing a fitted marginal model.
id	An integer or factor vector identifying the clusters. Its length and ordering must match the number and ordering of the observations used to fit the model.
verbose	Logical. If TRUE, the value of the RJC is printed to the console.

Details

The RJC is defined using the sensitivity and variability structures of the estimating equations and measures the discrepancy between them. The implementation assumes that the data are correctly ordered such that observations belonging to the same cluster are stored in contiguous rows.

Warning: This function is restricted to models with a single response variable.

Value

An invisible list with a single component:

RJC

A numeric scalar giving the value of the Rotnitzky–Jewell information criterion.

Author(s)

Wagner Hugo Bonat, [e-mail]

Source

Wang, M. (2014). Generalized estimating equations in longitudinal data analysis: A review and recent developments. Advances in Statistics, 1(1), 1–13.

See Also

gof, plogLik, pAIC, pKLIC, ESS, GOSHO

mcglm documentation built on Jan. 9, 2026, 1:07 a.m.