A new estimation method for the semiparametric accelerated failure time mixture cure model
A new estimation method for the semiparametric accelerated failure time mixture cure model
复制标题
DOI：
10.1002/sim.2748
复制
发表时间：
2007-07-20
影响因子：
2
通讯作者：
Peng, Yingwei
中科院分区：
文献类型：
--
作者：
Zhang, Jiajia;Peng, Yingwei
比例风险(PH)混合物养护模型和加速失效时间(AFT)混合物养护模型常用于分析长期存活者的失效时间数据。然而，与半参数PH混合固化模型相比，半参数AFT混合固化模型由于其估计方法的复杂性而受到的关注较少。本文对半参数AFT混合曲线模型提出了一种新的估计方法。该方法利用EM算法和AFT模型的秩估计量来估计感兴趣的参数。EM算法中的M步结合了秩式估计方程，可以很容易地用线性规划方法来实现。为了评估该方法的性能，我们进行了仿真研究。仿真结果表明，该方法的性能优于现有的估计方法，半参数AFT混合固化模型比参数AFF混合固化模型提高了参数的可辨识性。为了说明这一点，我们将该模型和所提出的方法应用于骨髓移植患者失败次数的数据集。版权所有(C)2006 John Wiley&Sons，Ltd.
The proportional hazard (PH) mixture cure model and the accelerated failure time (AFT) mixture cure model are usually used in analysing failure time data with long-term survivors. However, the semiparametric AFT mixture cure model has attracted less attention than the semiparametric PH mixture cure model because of the complexity of its estimation method. In this paper, we propose a new estimation method for the semiparametric AFT mixture cure model. This method employs the EM algorithm and the rank estimator of the AFT model to estimate the parameters of interest. The M-step in the EM algorithm, which incorporates the rank-like estimating equation, can be carried out easily using the linear programming method. To evaluate the performance of the proposed method, we conduct a simulation study. The results of the simulation study demonstrate that the proposed method performs better than the existing estimation method and the semiparametric AFT mixture cure model improves the identifiability of the parameters in comparison to the parametric AFF mixture cure model. To illustrate, we apply the model and the proposed method to a data set of failure times from bone marrow transplant patients. Copyright (C) 2006 John Wiley & Sons, Ltd.