%0 Journal Article %1 Zhang2011 %A Zhang, Hui %A Lu, Naiji %A Feng, Changyong %A Thurston, Sally W %A Xia, Yinglin %A Zhu, Liang %A Tu, Xin M %D 2011 %J Statistics in medicine %K GLIMMIX integralapproximation linearization lm %N 20 %P 2562-2572 %R 10.1002/sim.4265 %T On fitting generalized linear mixed-effects models for binary responses using different statistical packages. %U http://dx.doi.org/10.1002/sim.4265 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3175267&tool=pmcentrez&rendertype=abstract %V 30 %X 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. %@ 1097-0258

@article{Zhang2011, 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.}, added-at = {2023-02-03T11:44:35.000+0100}, author = {Zhang, Hui and Lu, Naiji and Feng, Changyong and Thurston, Sally W and Xia, Yinglin and Zhu, Liang and Tu, Xin M}, biburl = {https://www.bibsonomy.org/bibtex/22280e8f65ba9005d4259effd0238c70b/jepcastel}, doi = {10.1002/sim.4265}, interhash = {9d4a6803eb161805144d82507aec938d}, intrahash = {2280e8f65ba9005d4259effd0238c70b}, isbn = {1097-0258}, issn = {1097-0258}, journal = {Statistics in medicine}, keywords = {GLIMMIX integralapproximation linearization lm}, month = {6}, note = 6181, number = 20, pages = {2562-2572}, pmid = {21671252}, timestamp = {2023-02-03T11:44:35.000+0100}, title = {On fitting generalized linear mixed-effects models for binary responses using different statistical packages.}, url = {http://dx.doi.org/10.1002/sim.4265 http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=3175267&tool=pmcentrez&rendertype=abstract}, volume = 30, year = 2011 }