线性模型中一类新的随机限制s-K估计  

A NEW STOCHASTIC RESTRICTED s-K ESTIMATOR IN THE LINEAR REGRESSION MODEL

作　　者：周玲[1] 徐静[1] ZHOU Ling XU Jing(Department of Statistics, Anhui Normal University, Wuhu 241003)

机构地区：[1]安徽师范大学统计系,芜湖241003

出　　处：《系统科学与数学》2016年第11期2118-2126,共9页Journal of Systems Science and Mathematical Sciences

基　　金：安徽省自然科学基金(1308085QA13)资助课题

摘　　要：研究带随机先验信息的线性回归模型的参数估计问题.为了克服复共线性的影响,通过融合s-K估计和普通混合估计,提出一类新的随机限制s-K估计.给出了新估计在均方误差阵意义下分别优于普通混合估计和s-K估计的充要条件以及调节参数的选择方法.最后,通过Monte Carlo模拟研究了新估计的表现.In this paper, the parameter estimation in linear regression model with additional stochastic linear restrictions is investigated. To overcome the multicollinear- ity problem, a new stochastic restricted s-K estimator is introduced by combining the 8-K estimator and the ordinary mixed estimator （OME）. Then necessary and suffi- cient conditions for the superiority of the new estimator over the OME and the s-K estimator are derived by the mean squared error matrix criterion. And a selection method for regulation parameters is also given. Finally, a Monte Carlo simulation study is carried out to investigate the performance of the proposed estimator.

关 键 词：复共线性 普通混合估计 随机限制s-K估计 均方误差阵 

分 类 号：O212.1[理学—概率论与数理统计]