Lou Yichen, Du Mingyue
School of Physical and Mathematical Sciences, Nanyang Technological University, Singapore.
School of Mathematics, 12510 Jilin University , Changchun, China.
Int J Biostat. 2025 Aug 29. doi: 10.1515/ijb-2024-0016.
This paper discusses regression analysis of interval-censored failure time data arising from semiparametric transformation models in the presence of covariates that are missing at random (MAR). We define a specific formulation of the MAR mechanism tailored to the interval censoring, where the timing of observation adds complexity to handling missing covariates. To overcome the limitations and computational challenges present in the existing methods, we propose a multiple imputation procedure that can be easily implemented with the use of the standard software. The proposed method makes use of two predictive scores for each individual and the distance defined by these scores. Furthermore, it utilizes partial information from incomplete observations and thus yields more efficient estimators than the complete-case analysis and the inverse probability weighting approach. An extensive simulation study is conducted to assess the performance of the proposed method and indicates that it performs well in practical situations. Finally we apply the proposed approach to an Alzheimer's Disease study that motivated this work.
本文讨论了在存在随机缺失协变量（MAR）的情况下，由半参数变换模型产生的区间删失失效时间数据的回归分析。我们定义了一种专门针对区间删失的MAR机制的具体形式，其中观测时间增加了处理缺失协变量的复杂性。为了克服现有方法存在的局限性和计算挑战，我们提出了一种多重填补程序，该程序可以使用标准软件轻松实现。所提出的方法利用每个个体的两个预测得分以及由这些得分定义的距离。此外，它利用了来自不完整观测的部分信息，因此比完整病例分析和逆概率加权方法产生更有效的估计量。进行了广泛的模拟研究以评估所提出方法的性能，结果表明该方法在实际情况下表现良好。最后，我们将所提出的方法应用于一项激发此项工作的阿尔茨海默病研究。
