Research Article

On fitting generalized linear mixed-effects models for binary responses using different statistical packages

  1. Hui Zhang1,*,
  2. Naiji Lu2,3,
  3. Changyong Feng2,
  4. Sally W. Thurston2,
  5. Yinglin Xia2,3,
  6. Liang Zhu1,
  7. Xin M. Tu2,3,4

Article first published online: 10 JUN 2011

DOI: 10.1002/sim.4265

Issue

Statistics in Medicine

Statistics in Medicine

Volume 30, Issue 20, pages 2562–2572, 10 September 2011

Additional Information

How to Cite

Zhang, H., Lu, N., Feng, C., Thurston, S. W., Xia, Y., Zhu, L. and Tu, X. M. (2011), On fitting generalized linear mixed-effects models for binary responses using different statistical packages. Statist. Med., 30: 2562–2572. doi: 10.1002/sim.4265

Author Information

  1. 1

    Department of Biostatistics, St. Jude Children's Research Hospital, Memphis, TN 38105, U.S.A.

  2. 2

    Department of Biostatistics and Computational Biology, University of Rochester, Rochester, NY 14642, U.S.A.

  3. 3

    Department of Psychiatry, University of Rochester, Rochester, NY 14642, U.S.A.

  4. 4

    VA Center of Excellence at Canandaigua, Canandaigua VAMC, Canandaigua, NY 14424, U.S.A.

*Department of Biostatistics, St. Jude Children's Research Hospital, Memphis, TN 38105, U.S.A.

Publication History

  1. Issue published online: 17 AUG 2011
  2. Article first published online: 10 JUN 2011
  3. Manuscript Accepted: 21 MAR 2011
  4. Manuscript Received: 30 OCT 2009

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

View Full Article (HTML) Get PDF (142K)

Keywords:

  • integral approximation;
  • linearization;
  • GLIMMIX;
  • lme4;
  • NLMIXED;
  • R;
  • SAS;
  • ZELIG

Abstract

The generalized linear mixed-effects model (GLMM) is a popular paradigm to extend models for cross-sectional data to a longitudinal setting. When applied to modeling binary responses, different software packages and even different procedures within a package may give quite different results. In this report, we describe the statistical approaches that underlie these different procedures and discuss their strengths and weaknesses when applied to fit correlated binary responses. We then illustrate these considerations by applying these procedures implemented in some popular software packages to simulated and real study data. Our simulation results indicate a lack of reliability for most of the procedures considered, which carries significant implications for applying such popular software packages in practice. Copyright © 2011 John Wiley & Sons, Ltd.

View Full Article (HTML) Get PDF (142K)