Working correlation structure selection in generalized estimating equations
- 137 Downloads
Selecting an appropriate correlation structure in analyzing longitudinal data can greatly improve the efficiency of parameter estimation, which leads to more reliable statistical inference. A number of such criteria have been proposed in the literature from different perspectives. However, little is known about the relative performance of these criteria. We review and evaluate these criteria by carrying out extensive simulation studies. Surprisingly, we find that the AIC and the BIC based on either the Gaussian working likelihood or the empirical likelihood outperform the others.
KeywordsCorrelation information criterion Empirical likelihood Longitudinal data Model selection
This research was funded by the Australian Research Council Discovery Projects (DP130100766 and DP160104292). L. Fu’s research was partly supported by the National Science Foundation of China (Grant Nos. 11201365 and 11301408) and the Doctoral Programs Foundation of Ministry of Education of China (Grant No. 2012020112005)and the Fundamental Research Funds for the Central Universities (Grant No. xjj2017180).
- Davidson R, MacKinnon JG (2004) Econometric theory and methods. Oxford University Press, OxfordGoogle Scholar
- Erhardt V (2009) Generate correlated count random variables. R package version 1.4. https://cran.rproject.org/web/packages/corcounts/index.html
- Genz A, Bretz F, Miwa T, Mi X, Leisch F, Scheipl F, Bornkamp B, Hothorn T (2014) Multivariate normal and t distributions. R package version 0.9-9997. http://CRAN.R-project.org/package=bindata
- Leisch F, Weingessel A, Hornik K (2011) Generation of artificial binary data. R package version 0.9-19. http://CRAN.R-project.org/package=bindata