search for




 

Accelerated failure time models for right censored failure time data
Journal of the Korean Data & Information Science Society 2018;29:1365-79
Published online November 30, 2018
© 2018 Korean Data and Information Science Society.

Sangwook Kang1 · Kyu hyun Kim2

12Department of Applied Statistics, Yonsei University
Correspondence to: Graduate Student, Department of Applied Statistics, Seoul 03722, Korea. E-mail: kkh@yonsei.ac.kr
This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government(MEST) (No. 2017R1A2B4005818).
Received October 10, 2018; Revised November 16, 2018; Accepted November 16, 2018.
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
Semiparametric accelerated failure time model postulates a linear relationship between the log-transformed failure time and a set of covariates. Since covariates work directly on the failure time, the effects of covariates are easy to understand. Despite these merits, the model has been less popular than the Cox model in analyzing rightcensored failure time data. It is mainly due to the inefficiency and instability in calculating model parameter estimates and lack of computer software implementing these methods. Since the mid 2000, these problems have been dramatically reduced with the emergence of some promising methods. In this paper, we review recently developed statistical inference procedures for accelerated failure time model with focusing on semiparametric accelerated failure time model.
Keywords : Induced smoothing, least-squares method, rank-based estimation, resampling, right-censoring.