Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data

Interval censored data arise from many clinical studies when the failure event cannot be directly observed. Methods developed in the literature for semiparametric regression analysis of such data are mainly based on common survival models, such as the proportional hazards, proportional odds and acce...

Full description

Saved in:
Bibliographic Details
Main Authors: Szabo, Zsolt, Liu, Xiaoyu, Xiang, Liming
Other Authors: School of Physical and Mathematical Sciences
Format: Article
Language:English
Published: 2022
Subjects:
Online Access:https://hdl.handle.net/10356/159515
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-159515
record_format dspace
spelling sg-ntu-dr.10356-1595152022-06-27T02:51:56Z Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data Szabo, Zsolt Liu, Xiaoyu Xiang, Liming School of Physical and Mathematical Sciences Science::Mathematics Asymptotics Splines Interval censored data arise from many clinical studies when the failure event cannot be directly observed. Methods developed in the literature for semiparametric regression analysis of such data are mainly based on common survival models, such as the proportional hazards, proportional odds and accelerated failure time models. As an alternative, the accelerated hazards (AH) model is of great practical interest for modeling survival data in some situations for which those conventional models would be inappropriate. However, inference in the AH model is particularly challenging due to the fact that the stochastic order of the observations depends on the finite dimensional parameters in the model. In this paper, we propose a sieve maximum likelihood estimation procedure based on polynomial splines for the AH model with interval censored data and develop a two-step maximization algorithm for its implementation. We establish large sample properties of the resulting estimator and evaluate its finite sample performance through simulation studies. For illustration purpose, the proposed method is applied to analysis of diabetes conversion data collected from an intervention study of individuals at high risk of developing diabetes. Ministry of Education (MOE) This work was supported in part by the Singapore Ministry of Education Academic Research Fund Tier 2 grant (MOE2013-T2-2-118) and Tier 1 grant (RG134/17 (S)). 2022-06-27T02:51:56Z 2022-06-27T02:51:56Z 2020 Journal Article Szabo, Z., Liu, X. & Xiang, L. (2020). Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data. Journal of Statistical Planning and Inference, 205, 175-192. https://dx.doi.org/10.1016/j.jspi.2019.07.002 0378-3758 https://hdl.handle.net/10356/159515 10.1016/j.jspi.2019.07.002 2-s2.0-85069727459 205 175 192 en MOE2013-T2-2-118 RG134/17 (S) Journal of Statistical Planning and Inference © 2019 Elsevier B.V. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Science::Mathematics
Asymptotics
Splines
spellingShingle Science::Mathematics
Asymptotics
Splines
Szabo, Zsolt
Liu, Xiaoyu
Xiang, Liming
Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
description Interval censored data arise from many clinical studies when the failure event cannot be directly observed. Methods developed in the literature for semiparametric regression analysis of such data are mainly based on common survival models, such as the proportional hazards, proportional odds and accelerated failure time models. As an alternative, the accelerated hazards (AH) model is of great practical interest for modeling survival data in some situations for which those conventional models would be inappropriate. However, inference in the AH model is particularly challenging due to the fact that the stochastic order of the observations depends on the finite dimensional parameters in the model. In this paper, we propose a sieve maximum likelihood estimation procedure based on polynomial splines for the AH model with interval censored data and develop a two-step maximization algorithm for its implementation. We establish large sample properties of the resulting estimator and evaluate its finite sample performance through simulation studies. For illustration purpose, the proposed method is applied to analysis of diabetes conversion data collected from an intervention study of individuals at high risk of developing diabetes.
author2 School of Physical and Mathematical Sciences
author_facet School of Physical and Mathematical Sciences
Szabo, Zsolt
Liu, Xiaoyu
Xiang, Liming
format Article
author Szabo, Zsolt
Liu, Xiaoyu
Xiang, Liming
author_sort Szabo, Zsolt
title Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
title_short Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
title_full Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
title_fullStr Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
title_full_unstemmed Semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
title_sort semiparametric sieve maximum likelihood estimation for accelerated hazards model with interval-censored data
publishDate 2022
url https://hdl.handle.net/10356/159515
_version_ 1736856398987264000