摘要
case-cohort抽样作为一种具有成本效益的有偏抽样设计,在多个领域得到应用.该文将case-cohort抽样后得到的数据应用于比例风险模型中,通过使用逆概率加权方法修正在case-cohort抽样中协变量信息观测不完全且补充样本的分布与总体的分布不相同带来的影响,并证明了逆概率加权后参数估计值的渐近性质.最后,通过Newton-Raphson迭代算法进行数值模拟,进一步说明了case-cohort抽样下比例风险模型中参数的逆概率加权估计方法在理论中的价值.
As a cost-effective biased sampling design,case-cohort sampling has been applied in many fields,which has certain theoretical significance and application value.In this paper,the data obtained from case-cohort sampling are applied to the proportional hazards model.By using the inverse probability weighting method,the influence of incomplete observation of covariate information and different distribution of supplementary sample and population in case-cohort sampling is corrected.The asymptotic property and proof process of parameter estimation after inverse probability weighting are given.Finally,the Newton-Raphson iterative algorithm is used for numerical simulation to further illustrate the application value of the inverse probability weighted estimation method of parameters in proportional hazards model under case-cohort sampling in theory.
作者
陈丽玲
韦程东
唐璐薇
罗文婷
林玉婷
周昭君
CHEN Li-ling;WEI Cheng-dong;TANG Lu-wei;LUO Wen-ting;LIN Yu-ting;ZHOU Zhao-jun(School of Mathematics and Statistics,Nanning Normal University,Nanning 530100,China)
出处
《南宁师范大学学报(自然科学版)》
2021年第2期54-61,共8页
Journal of Nanning Normal University:Natural Science Edition
基金
国家自然科学基金(11561010)。