CFE 2019: Start Registration
View Submission - CMStatistics
Title: Smoothed time-dependent ROC curve for right censored survival data Authors:  Kassu Mehari Beyene - UCLouvain (Belgium) [presenting]
Anouar El Ghouch - The University catholique de Louvain (Belgium)
Abstract: The prediction reliability is of primary concern in many clinical studies when the objective is to develop new predictive models. In fact, prior using a model in any clinical decision making, it is very important to check its ability to discriminate between subjects who are in risk of developing certain disease from those who will not. To that end, the time-dependent receiver operating characteristic curve (ROC) is the most commonly used method in practice. Several approaches have been proposed in the literature to estimate the ROC non-parametrically in the context of survival data. But, except one recent approach, all the existing methods provide a non-smooth ROC estimator whereas, by definition, the ROC curve is smooth. We propose and study a new non-parametric smooth ROC estimator based on a weighted kernel smoothing. As bandwidth is the main parameter to be set, we present and study different methods to appropriately select one. A simulation study is conducted, under different scenarios, to prove the consistency of the proposed method and compare its finite sample performance with a competitor. The results show that the proposed method performs better. Furthermore, we illustrate the method using a real data example.