Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 79 - Contributed Poster Presentations: Lifetime Data Science Section
Type: Contributed
Date/Time: Monday, August 3, 2020 : 10:00 AM to 2:00 PM
Sponsor: Lifetime Data Science Section
Abstract #312993
Title: Discrete Time Survival Analysis with Dependent Censoring
Author(s): Jung Ae Lee*
Companies: University of Arkansas Agriculture Statistics Laboratory
Keywords: survival analysis ; missing data; dependent censoring; cox regression; discrete time; repeated measures
Abstract:

The main strength of survival analysis such as Kaplan-Meier estimate or Cox regression is their ability to handle missing values (e.g., drop-outs), so called a right-censored data. To provide unbiased results under censoring, conditional independence between event time and censoring should be assumed. Independent censoring, however, is untestable assumption while it is still questionable in reality. Therefore, there have been many studies to examine the influence of dependence censoring in a variety of format, such as sensitivity analysis using copula. The sensitivity studies can be performed on several aspects such as in the estimation of coefficient, percent of censoring, statistical power, sample size, prediction and many other important statistics. As part of this effort, this study focuses on experimental data that are repeatedly measured on discrete time points, in which we expect many tie values on response. Alternative linear models for longitudinal data will be explored regarding the missing data handling.


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

Back to the full JSM 2020 program