带有治愈亚组和错误分类的当前状态数据的回归分析

Regression Analysis of Misclassified Current Status Data with A Cured Subgroup

带有治愈亚组和错误分类的当前状态数据常出现在流行病学调查和医学临床试验等科学领域中。这类数据通常具有三个复杂特点,首先,我们对每个研究个体只检测一次以确定感兴趣事件的发生状态,对于感兴趣事件的发生时间,只知道其大于或小于检测时间,而非被精确观测到;其次,用于确定感兴趣事件发生状态的检测仪器通常具有一定的检测错误,由此产生带有错误分类的数据;此外,研究中有一部分个体可能永远不会发生感兴趣的事件,因此研究总体中可能存在治愈亚组。本文主要讨论带有治愈亚组和错误分类的当前状态数据的回归分析问题。特别地,我们假定感兴趣事件的发生时间服从一类半参数转移非混合治愈模型,提出一种期望极大化(EM)算法来极大化具有复杂形式的观测数据似然函数以得到参数估计。数值模拟结果表明所提出的估计方法在有限样本下具有良好的表现,并且优于忽略错误分类的naive方法。我们还将所提出的方法应用到一组有关于衣原体感染的实际数据中。

Misclassified current status data with a cured subgroup are commonly encountered in various scientific areas including epidemiological studies and clinical trials in which three complicated features often exist.Firstly,each subject under study is observed only once to determine the failure status of the event of interest,and the failure time of interest is only known to be smaller or larger than the examination time rather than observed exactly;Secondly,the diagnostic test used to determine the failure status of the event may not be perfect,which yields misclassified data in practice;Thirdly,a proportion of subjects under study may never experience the event of interest,and thus we have a cured subgroup in the whole population.In this paper,we discuss regression analysis of misclassified current status data with a cured subgroup.In particular,we assume the event time of interest follows a wide class of semiparametric transformation nonmixture cure model,and to obtain the parameter estimators,an expectation-maximization(EM)algorithm is developed to maximize the observed data likelihood function with complex form.The numerical results shown in simulation study indicate that the proposed approach performs well and is superior to the naive method that ignores the misclassification.We also apply the proposed methodology to a set of real data on chlamydia.