geeCRT
CRANgeeCRTMaRDI QIDQ71336FDOQ71336
Bias-Corrected GEE for Cluster Randomized Trials
Fan Li, Hengshi Yu, John Preisser, Paul Rathouz, Elizabeth L. Turner
Last update: 19 February 2024
Copyright license: GNU General Public License, version 3.0, GNU General Public License, version 2.0
Software version identifier: 1.1.0, 0.0.1, 0.1.0, 0.1.1, 1.1.1, 1.1.2, 1.1.3
Population-averaged models have been increasingly used in the design and analysis of cluster randomized trials (CRTs). To facilitate the applications of population-averaged models in CRTs, the package implements the generalized estimating equations (GEE) and matrix-adjusted estimating equations (MAEE) approaches to jointly estimate the marginal mean models correlation models both for general CRTs and stepped wedge CRTs. Despite the general GEE/MAEE approach, the package also implements a fast cluster-period GEE method by Li et al. (2022) <doi:10.1093/biostatistics/kxaa056> specifically for stepped wedge CRTs with large and variable cluster-period sizes and gives a simple and efficient estimating equations approach based on the cluster-period means to estimate the intervention effects as well as correlation parameters. In addition, the package also provides functions for generating correlated binary data with specific mean vector and correlation matrix based on the multivariate probit method in Emrich and Piedmonte (1991) <doi:10.1080/00031305.1991.10475828> or the conditional linear family method in Qaqish (2003) <doi:10.1093/biomet/90.2.455>.
- Marginal modeling of cluster-period means and intraclass correlations in stepped wedge designs with binary outcomes
- A Method for Generating High-Dimensional Multivariate Binary Variates
- A family of multivariate binary distributions for simulating correlated binary variables with specified marginal means and correlations
This page was built for software: geeCRT
