半参数回归模型在面板计数数据中的最大似然估计。
Maximum Likelihood Estimation for Semiparametric Regression Models With Panel Count Data.
发表日期：2021 Dec
作者：
By Donglin Zeng, D Y Lin
来源：
BIOMETRIKA
摘要：
面板计数数据，其中每个研究对象的观察结果由连续检查之间出现的再发事件次数组成，在工业可靠性测试、医学研究和其他各种科学研究中常见。我们通过具有随机效应的非齐性泊松过程来制定潜在时间依赖协变量对一种或多种经常性事件的影响。在任意检查方案下采用非参数最大似然估计，并发展了一个简单而稳定的EM算法。我们证明回归参数的估计值是一致的和渐近正态的，其协方差矩阵达到半参数效率界，并且可以通过概率轮廓似然估计。我们通过大量模拟研究评估所提出方法的性能，并提供皮肤癌临床试验。
Panel count data, in which the observation for each study subject consists of the number of recurrent events between successive examinations, are commonly encountered in industrial reliability testing, medical research, and various other scientific investigations. We formulate the effects of potentially time-dependent covariates on one or more types of recurrent events through non-homogeneous Poisson processes with random effects. We adopt nonparametric maximum likelihood estimation under arbitrary examination schemes and develop a simple and stable EM algorithm. We show that the resulting estimators of the regression parameters are consistent and asymptotically normal, with a covariance matrix that achieves the semiparametric efficiency bound and can be estimated through profile likelihood. We evaluate the performance of the proposed methods through extensive simulation studies and present a skin cancer clinical trial.