|
Activity Number:
|
353
- SPEED: Statistical Learning and Data Science Speed Session 2, Part 2
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 11:15 AM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #307732
|
|
|
Title:
|
An Imputation Approach for Fitting Random Survival Forests with Interval-Censored Survival Data
|
|
Author(s):
|
Warren Keil* and Tyler Cook
|
|
Companies:
|
and University of Central Oklahoma
|
|
Keywords:
|
Survival Analysis;
Random Forest;
Interval Censoring;
Imputation
|
|
Abstract:
|
Recently, much work has focused on extending machine learning methods to handle survival data. Currently, the majority of the research in this area focuses on data with right censoring. Few methods exist for interval-censored data even though this censoring type is a general class of survival data commonly seen in medical studies with periodic follow-up. Here, we propose a novel approach for fitting random survival forests with interval-censored data using multiple imputation. Imputation procedures have proven effective for a variety of problems with interval-censored data and they often have the benefit of reducing the problem to the easier-to-handle case of right censoring. This new method allows for flexible estimation of the survival function and easily produces variable importance scores. Through extensive simulation studies and an analysis of a real dataset, we show that this imputed random survival forest compares favorably with other existing methods for the analysis of interval-censored data.
|
Authors who are presenting talks have a * after their name.
