基于病例队列数据的比例风险模型的诊断

病例队列设计是一种在生存分析中广泛应用的可以降低成本又能提高效率的抽样方法.对于病例队列数据,已经有很多统计方法基于比例风险模型来估计协变量对生存时间的影响.然而,很少有工作基于病例队列数据来检验模型的假设是否成立.在这篇文章中,我们基于渐近的零均的值随机过程提出了一类检验统计量,这类检验统计量可以基于病例队列数据来检验比例风险模型的假设是否成立.我们通过重抽样的方法来逼近上述检验统计量的渐近分布,通过数值模拟来研究所提方法在有限样本下的表现,最后将所提出的方法应用于一个国家肾母细胞瘤研究的真实数据集上.

Model Diagnostics for the Proportional Hazards Model with Case-Cohort Data

Case-cohort design is a well-known cost-effective design and has been widely used in survival analysis. Many statistical methods have been developed to estimate the covariates effects on the survival time based on case-cohort data. However, little work has focused on checking the proportional hazards model assumptions with case-cohort data. In this article, we propose a class of test statistics through the asymptotically mean-zero processes for testing the proportional hazards assumption with case-cohort data. Re-sampling scheme is proposed to approximate the asymptotic distribution of the test statistics. Simulation studies are conducted to evaluate the finite sample performances of the proposed method and a data set from the National Wilm's Tumor Study Group is analyzed to illustrate the proposed method.