变系数部分线性模型的加权随机约束s-K估计

研究随机约束条件下半参数变系数部分线性模型的参数估计问题，当回归模型线性部分变量存在多重共线性时，基于Profile最小二乘方法、s-K估计和加权混合估计构造参数向量的加权随机约束s-K估计，并在均方误差矩阵准则下给出新估计量优于s-K估计和加权混合估计的充要条件，最后通过蒙特卡洛数值模拟验证所提出估计量的有限样本性质.

A Weighted Stochastic Restricted s-K Estimator in Varying-Coefficient Partially Linear Model

The parameter estimation problem of semiparametric varying-coefficient partially linear model is studied when the linear part variables of the regression model exist multicollinearity. Based on the stochastic linear constraint, a weighted stochastic restricted s-K estimator for parameter vector is constructed by using the profile least squares method, the s-K estimation and the weighted mixed estimation. The necessary and sufficient conditions for the new estimator to be superior to the weighted mixed estimator and the s-K estimator are given out under the scalar mean square error matrix criterion. Finally, the finite sample properties of the proposed estimators are verified by Monte Carlo numerical simulation.