eduzhai > Physical Sciences > Physics Sciences >

A Comparative Analysis of Generalized Estimating Equations Methods for Incomplete Longitudinal Ordinal Data with Ignorable Dropouts

  • Save

... pages left unread,continue reading

Document pages: 23 pages

Abstract: In longitudinal studies,measurements are taken repeatedly over time on the same experimental unit.These measurements are thus correlated. Missing data are very common inlongitudinal studies. A lot of research has been going on ways to appropriatelyanalyze such data set. Generalized Estimating Equations (GEE) is a popularmethod for the analysis of non-Gaussian longitudinal data. In the presence ofmissing data, GEE requires the strong assumption of missing completely atrandom (MCAR). Multiple Imputation Generalized Estimating Equations (MIGEE),Inverse Probability Weighted Generalized Estimating Equations (IPWGEE) andDouble Robust Generalized Estimating Equations (DRGEE) have been proposed aselegant ways to ensure validity of the inference under missing at random (MAR).In this study, the three extensions of GEE are compared under various dropout ratesand sample sizes through simulation studies. Under MAR and MCAR mechanism, thesimulation results revealed better performance of DRGEE compared to IPWGEE andMIGEE. The optimum method was applied to real data set.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×