Abstract Details
Activity Number:
|
380
|
Type:
|
Topic Contributed
|
Date/Time:
|
Tuesday, August 5, 2014 : 2:00 PM to 3:50 PM
|
Sponsor:
|
IMS
|
Abstract #311894
|
View Presentation
|
Title:
|
Jackknife Empirical Likelihood-Based Variable Selection for Accelerated Failure Time Models
|
Author(s):
|
Xuewen Lu*+ and Longlong Huang and Karen Kopciuk
|
Companies:
|
University of Calgary and University of Calgary and Alberta Health Services/University of Calgary
|
Keywords:
|
Accelerated failure time model ;
Censored data ;
Jackknife empirical likelihood ;
Rank regression ;
U-statistics ;
Variable selection
|
Abstract:
|
Empirical likelihood based variable selection has been found to be very useful in many different settings. However, when it is used with more complicated statistics such as U-statistics, which are common in rank estimation methods for survival data, its theoretical properties are unknown. Substantial computational difficulties also frequently arise. Jackknife empirical likelihood (JEL) method has been shown to be very effective in handling U-statistics for many types of data. We propose an adjusted JEL based variable selection method for semiparametric accelerated failure time models with censored survival data and develop a new algorithm for its implementation. Simulation study results demonstrate that the new method is robust to model misspecification and is comparable in accuracy of variable selection to the classical information criteria based methods for parametric survival models. Theoretical results extend the scope of current knowledge about variable selection in semiparametric models for survival data analysis, where standard approaches fail due to model complexity.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2014 program
|
2014 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Professional Development program, please contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Copyright © American Statistical Association.