Multiply Robust Weighted Generalized Estimating Equations for Incomplete Longitudinal Binary Data Using Empirical Likelihood

In clinical trials, missing data may lead to serious misinterpretation of trial results. To address this issue, it is important collect post-randomization (such as efficacy measurement and adverse event onset data). Such are called auxiliary variables they can be useful for constructing missingness imputation models. A multiply robust estimator using an empirical likelihood method was previously proposed by Han Wang Han. However, that developed cross-sectional situations in which no missing. This contrary actual settings, some will invariably Consequently, apply Han’s longitudinal data, need imputed. article proposes a new extends outcome model applying weighted generalized estimating equations with weights. Monte Carlo simulations repeated binary response at random dropouts demonstrated the exhibits better performance than augmented inverse probability complete-case under several simulation scenarios. We also successfully applied plaque psoriasis study data.