For decades, much research has been devoted to developing and comparing variable selection methods, but primarily for the classical case of independent observations. Existing variable-selection methods can be adapted to cluster-correlated observations, but some adaptation is required. For example, classical model fit statistics such as AIC and BIC are undefined if the likelihood function is unk...

INTRODUCTION The objective of this study was to determine whether perceptions of the home and school food environments are related to food and beverage intakes of postpartum teens. METHODS Our study was a baseline, cross-sectional analysis of 853 postpartum teens enrolled in a weight-loss intervention study across 27 states from 2007 through 2009. Eight-item scales assessed perceived accessib...

The logistic Generalized Estimating Equations (logisticGEE) models have been extensively used for analyzing clustered binary data. However, assessing the goodness-of-fit and predictability of these models is problematic due to the fact that no likelihood is available and the observations can be correlated within a cluster. In this paper we propose a new measure for estimating the generalization...

The Generalized Estimating Equations GEE proposed by Liang and Zeger have found considerable attention in the last years and sev eral extensions have been proposed This paper will give a more intuitive description how GEE have been developed during the last years Addi tionally we will describe the advantages and disadvantages of the di erent parametrisations that have been proposed in the liter...

The Generalized Estimating Equations (GEE) proposed by Liang and Zeger (1986) have found considerable attention in the last years and several extensions have been proposed. This paper will give a more intuitive description how GEE have been developed during the last years. Additionally we will describe the advantages and disadvantages of the diierent parametrisations that have been proposed in ...

We consider marginal semiparametric partially linear models for longitudinal/clustered data and propose an estimation procedure based on a spline approximation of the nonparametric part of the model and an extension of the parametric marginal generalized estimating equations (GEE). Our estimates of both parametric part and nonparametric part of the model have properties parallel to those of par...