基于随机效应模型的纵向计数数据统计分析

Statistical Analysis of Longitudinal Count Data Based on the Random Effects Model

文章对复发事件过程中的纵向计数数据进行研究.充分考虑了纵向计数数据在重复观测过程中数据之间相关联的情况,建立纵向计数数据的随机效应模型.采用极大似然估计的方法对模型中的参数进行了估计,运过大样本理论,论证了估计量的相合性和渐近正态性.通过数值模拟的方法,量化地说明这些估计在有限样本量下的合理性.最后,结合真实的纵向计数数据实例,运用随机效应模型对这些数据进行统计分析.

The longitudinal count data arising from recurrent event is studied. By fully considering the dependence of the data between the each longitudinal count data in the repeated observation process,we consider modering longitudinal count data with random effects model. The maximum likelihood estimation is used to estimate the parameters in the random effects model. By using the Large-Sample theory,we demonstrate the consistency and asymptotic normality of estimated value. Then with the numerical simulation method,quantified the rationality of these estimations under finite sample size. Finally, combined with the real longitudinal count data in specific examples, we perform analysis of these data with random effects model.