异方差模型中估计方差的一种新方法  

A New Method of Estimating the Variance in Heteroscedastic Model

作　　者：张晓琴[1] 郝红霞[1] 

机构地区：[1]山西大学数学科学学院,山西太原030006

出　　处：《中北大学学报（自然科学版）》2013年第5期481-484,共4页Journal of North University of China(Natural Science Edition)

基　　金：国家自然科学基金重点项目(71031006;41101440);山西省青年科技基金资助项目(201202105-1)

摘　　要：研究了经典回归模型中的异方差现象,讨论了该现象中的异方差估计这一热点问题,利用交表提出了一种不依赖于数据分布类型的计算简便的异方差估计方法,把这一方法应用在研究三大产业对GDP影响的一个三维数据的具体例子中,并与已有文献中得到的结果进行了比较.比较结果表明:利用正交表的方法对异方差进行估计是一种相对有效的方法.The phenomenon of heteroscedasticity in classic linear regression model was investigated and a hot issue about the estimation of heteroscedastic was discussed. With the help of orthogonal array, a nonparamet- ric estimation method with simple calculation was proposed, which did not rely on the distribution of data. This method was applied to a specific case with data of three dimensions, which was studied the influence of three industries on the GDP and was compared with the result in the existed literature. Comparison of the results indicate that the estimation of heteroscedastic by orthogonal array is a relatively efficient way.

关 键 词：正交表 异方差 非参数估计 

分 类 号：O212.1[理学—概率论与数理统计]