Zhang Feipeng, Chen Xuerong, Zhou Yong
School of Statistics and Management, Shanghai University of Finance and Economics, Shanghai, China.
Lifetime Data Anal. 2014 Jan;20(1):132-57. doi: 10.1007/s10985-013-9257-5. Epub 2013 May 7.
Length-biased data arise in many important applications including epidemiological cohort studies, cancer prevention trials and studies of labor economics. Such data are also often subject to right censoring due to loss of follow-up or the end of study. In this paper, we consider a proportional hazards model with varying coefficients for right-censored and length-biased data, which is used to study the interact effect nonlinearly of covariates with an exposure variable. A local estimating equation method is proposed for the unknown coefficients and the intercept function in the model. The asymptotic properties of the proposed estimators are established by using the martingale theory and kernel smoothing techniques. Our simulation studies demonstrate that the proposed estimators have an excellent finite-sample performance. The Channing House data is analyzed to demonstrate the applications of the proposed method.
长度偏倚数据出现在许多重要应用中，包括流行病学队列研究、癌症预防试验和劳动经济学研究。由于随访缺失或研究结束，此类数据也常常受到右删失的影响。在本文中，我们考虑一个针对右删失和长度偏倚数据的具有可变系数的比例风险模型，该模型用于非线性地研究协变量与一个暴露变量的交互作用。针对模型中的未知系数和截距函数，提出了一种局部估计方程方法。利用鞅理论和核平滑技术建立了所提估计量的渐近性质。我们的模拟研究表明所提估计量具有出色的有限样本性能。通过分析钱宁楼数据来展示所提方法的应用。
