Chen Chyong-Mei, Lin Chih-Ching, Wu Chih-Cheng, Tsai Jia-Ren
Institute of Public Health, National Yang Ming Chiao Tung University, Taipei, Taiwan, R.O.C.
Division of Nephrology, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan, R.O.C.
J Appl Stat. 2024 Nov 22;52(7):1361-1380. doi: 10.1080/02664763.2024.2426015. eCollection 2025.
This paper proposes a mixture regression model for competing risks data, where the logistic regression model is specified for the marginal probabilities of the failure types and the mean residual lifetime (MRL) model is assumed for the failure time given the failure of interest. The estimating equations (EEs) are derived to infer the logistic regression and MRL model separately. We further consider the situation where the covariates are subject to measurement error. The presence of measurement error imposes extra challenges for the analysis of complex time-to-event data. By using the above EEs as the correction-amenable original estimating functions, we propose a corrected score estimation, which does not require specifying the distributions for unobserved error-prone covariates. The proposed estimators are shown to be consistent and asymptotically normally distributed. The performance of the method is investigated by intensive simulation studies and two real examples are presented to illustrate the proposed methods.
本文提出了一种用于竞争风险数据的混合回归模型，其中针对失效类型的边际概率指定了逻辑回归模型，并针对给定感兴趣失效情况下的失效时间假设了平均剩余寿命（MRL）模型。推导了估计方程（EEs）以分别推断逻辑回归和MRL模型。我们进一步考虑协变量存在测量误差的情况。测量误差的存在给复杂的事件发生时间数据的分析带来了额外挑战。通过将上述EEs用作可校正的原始估计函数，我们提出了一种校正得分估计，该估计不需要指定未观察到的易出错协变量的分布。所提出的估计量被证明是一致的且渐近正态分布。通过密集的模拟研究考察了该方法的性能，并给出了两个实际例子来说明所提出的方法。
