We consider a variety of tests for testing goodness–of–fit in a parametric Cox proportional hazards (PH) and accelerated failure time (AFT) model in the presence of Type–II right censoring. The testing procedures considered can be divided in two categories: an approach involving transforming the data to a complete sample and an approach using test statistics that can directly accommodate Type-II right censoring. The power of the proposed tests are compared through a Monte Carlo study for various scenarios. It is found that both approaches are useful for testing exponentiality if the censoring proportion in a data set is lower than 30%, but that it is recommended to use the approach that first transforms the sample to a complete sample when one encounters higher censoring proportions.

我们考虑了多种测试，以在存在II类右删失的情况下测试参数Cox比例风险（PH）和加速故障时间（AFT）模型的拟合优度。所考虑的测试程序可以分为两类：一种涉及将数据转换为完整样本的方法，另一种使用可以直接适应II型权限检查的测试统计数据的方法。通过蒙特卡洛研究针对各种情况对建议的测试功能进行了比较。我们发现，如果数据集中的检查比例低于30％，则这两种方法都可用于测试指数，但是建议使用一种方法，当一个方法遇到较高的检查比例时，首先将样本转换为完整的样本。