一种基于N-W估计的异方差估计方法  

A Method of Heteroscedasticity Estimation Based on N-W Estimation

作　　者：张晓琴 郭雅静 李顺勇[3] ZHANG Xiao-qin;GUO Ya-jing;LI Shun-yong(School of Statistics,Shanxi University of Finance and Economics,Taiyuan 030006;School of Science,Hebei University of Architecture,Zhangjiakou 075024;School of Mathematical Sciences,Shanxi University,Taiyuan 030006,China)

机构地区：[1]山西财经大学统计学院,山西太原030006 [2]河北建筑工程学院数理系,河北张家口075024 [3]山西大学数学科学学院,山西太原030006

出　　处：《统计学报》2020年第2期31-38,共8页Journal of Statistics

基　　金：国家社科基金一般项目(17BTJ010);山西省回国留学人员科研资助项目(2017-020,109653901002);山西省基础研究计划项目(201701D121004,201901D111320)

摘　　要：在异方差的构成形式未知时,可以利用非参数方法给出方差较为精确的估计,进而估计未知参数,并对模型做出有效预测和拟合。本文基于N-W估计,提出了一种新的异方差估计方法(KNW方法),通过模拟实验和实例分析将这一方法与已有的异方差估计方法进行比较,证明该方法是有效的。模拟实验结果表明,不同的核函数在方差的核估计及模型的预测和拟合方面,差异性较小。In the case where the composition form of heteroscedasticity is unknown,the non-parametric method can be used to estimate the variance and unknown parameters,so that the model can be effectively predicted and fitted.Based on N-W estimation,this paper proposes a new method of heteroscedasticity estimation--KNW.The simulation experiments and case analysis are used to compare the proposed method with the existing heteroscedasticity estimation method,and to verify the effectiveness of the proposed method.The simulation results show that the different kernel functions have less difference in the kernel estimation of variance,also the prediction and fitting of the model.

关 键 词：异方差 核估计 N-W估计 HCCMEs 协方差阵 

分 类 号：F272.1[经济管理—企业管理] O212[经济管理—国民经济]