纵向数据部分线性测量误差模型的二次推断函数估计

基于纵向数据部分线性测量误差模型,研究了模型中兴趣参数部分回归系数的估计问题.首先采用B样条方法逼近模型中的非参数函数,然后提出修正的二次推断函数(QIF)方法对模型中参数部分的回归系数进行估计,所提方法可以提高估计的效率. 在一定的正则条件下,证明了所得到的估计量具有相合性和渐近正态性. 最后,通过模拟研究和实例分析验证了所提出估计方法的有限大样本性质.

Quadratic Inference Function Estimation in Partially Linear Measurement Error Models with Longitudinal Data

In this paper, we focus on the estimationfor the marginal semiparametric partially linear models with longitudinaldata when some covariates are measured with additive errors. We firstapproximate the nonparametric part of the model based on the B-splinesand a new bias-corrected estimation procedure is proposed based on thequadratic inference functions (QIF). Under some regularity conditions,we show that the QIF estimator is consistent and asymptotically normal.Extensive simulation studies and an application are conducted to examinethe finite sample performance of the proposed estimation procedure.