多类型复发事件数据下一类半参数转移模型

A Class of Semiparametric Transformation Models with Multiple Types of Recurrence Event Data

复发事件数据频繁的出现在纵向研究中,基于生物医学中的多类型复发事件数据,提出了一类半参数转移模型,该模型包含了一些重要的半参数模型。同时,模型允许协变量具有加性和乘性的影响,且加性影响随时间而变化。利用广义估计方程的思想,对模型中未知参数和非参数函数进行了估计,并且证明了估计的相合性和渐近正态性。

Recurrent event data appear frequently in longitudinal studies.Based on the data of multiple types of recurrence events in biomedicine,a semi-parametric transfer model is proposed,which contains some important semiparametric models.At the same time,the model allows covariates to have additive and multiplicative effects,and additive effects change over time.By using the idea of generalized estimating equations,the unknown and nonparametric functions in the model are estimated and the consistency and asymptotic normality of estimators are proved.