Semiparametric regression analysis of window-observation recurrent event data with multiple causes of failure

This article deals with the regression analysis of recurrent event data with multiple causes of failure that are collected in disconnected observation windows with gaps. These type of data are referred to as window-observation recurrence data and we introduce a proportional cause specific mean model for analyzing multiple causes of failure. We also develop methods for estimating the regression parameters and the baseline cause specific mean function. The asymptotic properties of the estimators are analyzed, and their performance is evaluated through simulation studies. The proposed techniques are then illustrated using a real data set.