一元线性模型异方差的局部多项式回归

Local Polynomial Regression for Heteroscedaticity in the Simple Linear Model

本文将局部多项式回归的非参数方法用于线性模型中异方差的估计 ,改进了传统的两阶段法 ,得到了估计的一致性和渐近正态性 ,为探讨估计的有限样本性 ,给出了若干模拟的例子。

We introduce the extension of the nonparametric regression technique of local polynomial fitting with a kernel weight to heteroscedastic linear regression model. One noteworthy feature of our approach is avoiding the testing for heteroscedasticity. We establish the uniform consistency and derive the asymptotic distribution. Simulation results show that our approach is effective in finite sample situations as well.