|国家预印本平台
首页|基于删失分类数据的AFT模型有效参数估计

基于删失分类数据的AFT模型有效参数估计

Efficient parameter estimation for the accelerated failure time model with clustered and censored data

中文摘要英文摘要

在医学研究中,搜集到的解释变量通常含有潜在的异常值。对于有删失的纵向数据/分类数据,传统的Gehan估计只对反应变量中的异常值稳健,但是对解释变量中的异常值敏感,且忽略了数据的相关性。考虑到数据的相关性,类的个数不同,和解释变量中的异常值,本文建议加权的Gehan估计函数来估计AFT模型中的参数,并证明了估计的渐近性质。最后通过数值模拟和实际数据分析,评估了建议方法的优良性。

In medical studies, the collected covariates usually contain underlying outliers. For clustered/longitudinal data withcensored observations, the traditional Gehan-type estimator is robust to outliers existing in response, but is sensitive to outliers in the covariate domain, and also ignores the within correlations. To take account of within correlations, varying cluster sizes, and outliers in covariates, this paper proposedweighted Gehan-type estimating functions to estimate parameters in the accelerated failure time model for clustered data, and prove theasymptotic properties of the resulting estimators. Furthermore, simulation studies were carried out to evaluate the performance ofthe proposed method. The simulation results demonstrate that the proposed method is robust to the outliers existing in the covariatedomain and lead to much more efficient estimators when a strong within correlation exists. Finally, the proposed method was usedupon a medical dataset and obtained a better result.

付利亚、 王友乾、周彦、 惠永昌

医学研究方法

删失数据稳健性分类数据

censored datarobustclustered data

付利亚, 王友乾,周彦, 惠永昌.基于删失分类数据的AFT模型有效参数估计[EB/OL].(2016-06-03)[2025-08-23].http://www.paper.edu.cn/releasepaper/content/201606-265.点此复制

评论