A0784
Title: Smoothed quantile regression For nonignorable missing longitudinal data
Authors: Tao Li - Shanghai University of Finance and Economics (China) [presenting]
Abstract: Quantile regression has become a popular tool for analyzing longitudinal data because it can characterize entire conditional distributions of responses and is robust to outliers. However, in real-world data, longitudinal outcomes often appear with nonignorable missing values. Several weighted estimators are proposed for quantile regression coefficients with nonignorable missing data based on parametric and semi-parametric missing mechanism models, respectively. To deal with the issue of excessive fluctuation of inverse probability weights, those stabilized weights are adopted. In addition, a smoothed empirical likelihood approach is developed to incorporate the within-subject correlations. It is shown that the proposed estimators are consistent and asymptotically normal. Simulation studies and two real-world examples are used to illustrate the usefulness of the proposed methods.
