Generalized estimating equations for analyzing multivariate survival data
Generalized estimating equations for analyzing multivariate survival data
复制标题
用于分析多变量生存数据的广义估计方程
DOI：
10.1080/03610918.2019.1619763
复制
发表时间：
2021
影响因子：
0.9
通讯作者：
Zhou Yan
中科院分区：
文献类型：
--
作者：
Fu Liya;Yang Zhuoran;Zhang Jun;Long Anle;Zhou Yan
广义估计方程 (GEE) 方法已用于估计具有聚类和审查数据的半参数加速失效时间 (AFT) 模型中的参数。然而，使用 GEE 方法时，工作相关模型对估计器效率有很大影响。本文提出了一种通用关联模型，用于合并聚类和审查数据之间的关联，并防止与错误指定的关联结构相关的可避免的效率损失。所提出的估计量是一致的且渐近正态的。进行仿真研究以证明所提出方法的有效性。最后，分析毒理学研究的真实数据集以进行说明。
Generalized estimating equations (GEE) approach has been used to estimate the parameters in semiparametric accelerated failure time (AFT) models with clustered and censored data. However, the working correlation model has a substantial impact on estimator efficiency when using the GEE method. This article proposes a general correlation model to incorporate the correlations among the clustered and censored data and protect against avoidable loss of efficiency associated with misspecified correlation structure. The proposed estimator is consistent and asymptotically normal. Simulation studies are carried out to demonstrate the effectiveness of the proposed method. Finally, a real dataset from a toxicology study is analyzed for illustration.