Comment on "Wang et al. (2005), Robust estimating functions and bias correction for longitudinal data analysis".

This note provides a discussion on the manuscript by Wang et al. (2005) who aim to robustify inference for longitudinal data analysis by replacing the ordinary generalized estimating function with an influence-bounded, possibly biased, version. To adjust for the bias of the ensuing robust estimator, the authors provide its analytic approximation by means of asymptotic expansions, and estimate it by plugging-in a nonrobust estimate of the parameter of interest. In this letter, we argue that the proposed bias-corrected estimator is, in fact, nonrobust.

Global sensitivity analysis for repeated measures studies with informative drop-out: A semi-parametric approach.

In practice, both testable and untestable assumptions are generally required to draw inference about the mean outcome measured at the final scheduled visit in a repeated measures study with drop-out. Scharfstein et al. (2014) proposed a sensitivity analysis methodology to determine the robustness of conclusions within a class of untestable assumptions. In their approach, the untestable and testable assumptions were guaranteed to be compatible; their testable assumptions were based on a fully parametric model for the distribution of the observable data. While convenient, these parametric assumptions have proven especially restrictive in empirical research. Here, we relax their distributional assumptions and provide a more flexible, semi-parametric approach. We illustrate our proposal in the context of a randomized trial for evaluating a treatment of schizoaffective disorder.