|
Activity Number:
|
634
- Dynamic Modeling for Timely Health Care Decisions
|
|
Type:
|
Invited
|
|
Date/Time:
|
Thursday, August 3, 2017 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
WNAR
|
|
Abstract #321882
|
View Presentation
|
|
Title:
|
Dynamic Predictions Based on Joint Models
|
|
Author(s):
|
Jeremy M. G. Taylor* and Krithika Suresh and Alex Tsodikov
|
|
Companies:
|
University of Michigan and University of Michigan and University of Michigan
|
|
Keywords:
|
dynamic prediction ;
joint models ;
landmarking
|
|
Abstract:
|
In this talk I will consider the problem of predicting a future event time for an individual based on his or her personal information that is available up to that point in time and using information that can be learned from a dataset of similar individuals. Joint models of longitudinal and survival data lend themselves naturally to this task, because from the joint distribution of the longitudinal and survival processes the required condition distribution can be derived. This approach to dynamic prediction will also satisfy the consistency conditions of Jewell and Nielsen (1993) for prediction models. In this talk I will present some examples of dynamic prediction based on joint models. I will also contrast this approach with those based on landmark analyses and partly conditional models which focus on using survival analysis techniques and avoid modeling the longitudinal data. These alternative methods do not satisfy the consistency condition, but may never-the-less be useful in particular situations if flexible models are carefully developed.
|
Authors who are presenting talks have a * after their name.
