Online Program Home
My Program

Abstract Details

Activity Number: 512 - Predicting and Evaluating Risk Models Within Distributions and Across Time
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 10:30 AM to 12:20 PM
Sponsor: Section on Risk Analysis
Abstract #306722
Title: A Simulation Approach to Predicting Time to Terminal Event in Joint Dynamic Modeling
Author(s): Piaomu Liu* and Edsel A Pena
Companies: Dept. of Mathematical Sciences, Bentley University and University of South Carolina
Keywords: Joint Dynamic Modeling; Recurrent Competing Risks; Terminal Event; Frailty ; Prediction; Simulation
Abstract:

In this short paper, we propose a simulation approach to predict time to terminal event (TE) that arises from joint dynamic modelling. Many joint dynamic models have found applications in medical research. Predicting terminal event times is often of the most interest due to its value as a prognostic tool in medical treatments and the complexity in developing appropriate prediction methodology. When a joint dynamic model gets more and more complicated, the computational aspect of predicting TE can be particularly challenging. An alternative, which is simulating censored event times according to history of data accrual has not been considered by previous works, to our best knowledge. Based on a class of joint dynamic models of recurrent competing risks (RCR) and terminal event (TE) in (Liu & Peña 2015), we demonstrate how to predict terminal event (TE) times by the simulation approach. We also point out the size-biased sampling related to gap time that traverses the monitoring time.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program