Online Program Home
My Program

Abstract Details

Activity Number: 184
Type: Contributed
Date/Time: Monday, August 1, 2016 : 10:30 AM to 12:20 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #320606
Title: Nonparametric Survival Analysis Using Bayesian Additive Regression Trees (BART)
Author(s): Rodney Sparapani* and Brent Logan and Robert McCulloch and Purushottam Laud
Companies: Medical College of Wisconsin and Medical College of Wisconsin and The University of Chicago and Medical College of Wisconsin
Keywords: Nonproportional hazards ; ensemble models ; predictive modeling

Bayesian additive regression trees (BART) provide a flexible framework for nonparametric modeling of covariate relationships with their outcomes. Recently, BART models have been shown to provide excellent predictive performance, for both continuous and binary outcomes, often exceeding that of its competitors and BART software is also readily available for this purpose. In this presentation, I will introduce extending the usefulness of BART in biomedical applications by addressing needs arising from survival analysis. Simulation studies of various regression scenarios, in comparison with long-standing traditional survival analysis methods, will establish the validity of this new approach. Using data from previously published study(ies), I will illustrate the use and some advantages of the proposed method in biomedical investigations.

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

Back to the full JSM 2016 program

Copyright © American Statistical Association