TY - JOUR
T1 - Empirical-likelihood-based criteria for model selection on marginal analysis of longitudinal data with dropout missingness
AU - Chen, Chixiang
AU - Shen, Biyi
AU - Zhang, Lijun
AU - Xue, Yuan
AU - Wang, Ming
N1 - Funding Information:
Wang's research was partially supported by Grant UL1 TR002014 and KL2 TR002015 from the National Center for Advancing Transnational Sciences (NCATS). The content is solely the responsibility of the authors and does not represent the official views of the National Institute of Health, the National Science Foundation, and other research sponsors.
Funding Information:
Wang's research was partially supported by Grant UL1 TR002014 and KL2 TR002015 from the National Center for Advancing Transnational Sciences (NCATS). The content is solely the responsibility of the authors and does not represent the official views of the National Institute of Health, the National Science Foundation, and other research sponsors.
Publisher Copyright:
© 2019 International Biometric Society
PY - 2019/9/1
Y1 - 2019/9/1
N2 - Longitudinal data are common in clinical trials and observational studies, where missing outcomes due to dropouts are always encountered. Under such context with the assumption of missing at random, the weighted generalized estimating equation (WGEE) approach is widely adopted for marginal analysis. Model selection on marginal mean regression is a crucial aspect of data analysis, and identifying an appropriate correlation structure for model fitting may also be of interest and importance. However, the existing information criteria for model selection in WGEE have limitations, such as separate criteria for the selection of marginal mean and correlation structures, unsatisfactory selection performance in small-sample setups, and so forth. In particular, there are few studies to develop joint information criteria for selection of both marginal mean and correlation structures. In this work, by embedding empirical likelihood into the WGEE framework, we propose two innovative information criteria named a joint empirical Akaike information criterion and a joint empirical Bayesian information criterion, which can simultaneously select the variables for marginal mean regression and also correlation structure. Through extensive simulation studies, these empirical-likelihood-based criteria exhibit robustness, flexibility, and outperformance compared to the other criteria including the weighted quasi-likelihood under the independence model criterion, the missing longitudinal information criterion, and the joint longitudinal information criterion. In addition, we provide a theoretical justification of our proposed criteria, and present two real data examples in practice for further illustration.
AB - Longitudinal data are common in clinical trials and observational studies, where missing outcomes due to dropouts are always encountered. Under such context with the assumption of missing at random, the weighted generalized estimating equation (WGEE) approach is widely adopted for marginal analysis. Model selection on marginal mean regression is a crucial aspect of data analysis, and identifying an appropriate correlation structure for model fitting may also be of interest and importance. However, the existing information criteria for model selection in WGEE have limitations, such as separate criteria for the selection of marginal mean and correlation structures, unsatisfactory selection performance in small-sample setups, and so forth. In particular, there are few studies to develop joint information criteria for selection of both marginal mean and correlation structures. In this work, by embedding empirical likelihood into the WGEE framework, we propose two innovative information criteria named a joint empirical Akaike information criterion and a joint empirical Bayesian information criterion, which can simultaneously select the variables for marginal mean regression and also correlation structure. Through extensive simulation studies, these empirical-likelihood-based criteria exhibit robustness, flexibility, and outperformance compared to the other criteria including the weighted quasi-likelihood under the independence model criterion, the missing longitudinal information criterion, and the joint longitudinal information criterion. In addition, we provide a theoretical justification of our proposed criteria, and present two real data examples in practice for further illustration.
UR - http://www.scopus.com/inward/record.url?scp=85072056373&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85072056373&partnerID=8YFLogxK
U2 - 10.1111/biom.13060
DO - 10.1111/biom.13060
M3 - Article
C2 - 31004449
AN - SCOPUS:85072056373
VL - 75
SP - 950
EP - 965
JO - Biometrics
JF - Biometrics
SN - 0006-341X
IS - 3
ER -