纵向可加部分线性测量误差模型的渐近估计

Asymptotic Estimation for Longitudinal Additive Partial Linear EV Models

基于纵向数据,研究参数部分协变量含有测量误差的可加部分线性测量误差模型的估计问题,提出了用于模型估计的偏差修正的二次推断函数方法,得到参数部分的估计结果具有相合性、渐近正态性,非参数可加函数的估计结果达到最优收敛速度。数值模拟和实例数据分析结果显示,该模型估计方法在同等条件下要优于广义估计方程方法。理论和数值结果显示,偏差修正的二次推断函数可以有效地处理测量误差和个体内相关性,是一个有效的纵向数据和测量误差数据分析工具,具有一定的理论和应用价值。

Based on longitudinal data,the estimation problem of the additive partial linear error-in-variables(EV)model is studied with measurement errors in parameter partial covariates,which proposes a bias-corrected quadratic inference functions for model estimation.The estimator of parametric part is consistent and asymptotic normality,and the estimator of nonparametric additive function convergence to the real function with attaining the optimal convergence rate.The results of numerical simulation and real data analysis show that,the proposed method is more proper.Theoretical and numerical results show that the bias-corrected quadratic inference functions method can effectively deal with measurement error and within-subject correlation,which is an effective tool for longitudinal data and measurement error data analysis.