Robust estimation for longitudinal data with informative observation times

Abstract

In this thesis, we focus on regression analysis of longitudinal data that often occur in medical follow-up studies and observational investigations. The analysis of these data involves two processes. One is the underlying recurrent event process of interest and the other is the observation process that controls observation times. Most of the existing methods, however, rely on some restrictive models or assumptions such as the Poisson assumption. For this, we propose a more general and robust estimation approach for regression analysis of longitudinal data with related observation times. The asymptotic properties of the proposed estimators are established and numerical studies indicate that the proposed method works well for practical situations.