摘要
所有的二维极值分布Ｌ（ｘ，ｙ）可以写成以下形式：Ｌ（ｘ，ｙ）＝ｅｘｐ｛－（ｅ－ｘ＋ｅ－ｙ）ｋ（ｙ－ｘ｜θ）｝其中ｋ（ｗ｜θ）称为相关函数，θ为参数常见的二维极值分布有四种类型，类型Ａ，Ｂ，Ｃ，Ｄ参数θ的估计方法通常采用象限法，本文对类型Ａ提出一种新的估计方法并证明该方法是无偏的。
All biextreme distributions L(x,y) can be written in the form L(x,y)=exp{-(e -x +e -y )k(y-x|θ)} where k(w|θ) is called the dependence function, θ the parameter. There are four types in the biextreme distributions which are commonly used, type A,B,C, ard D Quadrant method is generally used to estimate the parameters. In this paper, the estimation can be proved that the new method which is unbiased and relatively efficient.
出处
《天津理工学院学报》
1999年第A05期5-8,共4页
Journal of Tianjin Institute of Technology
关键词
相关函数
象限法
边缘分布
极值分布
参数估计
dependence functions
quadrant method
marginal distribution
Monte-Carlo Simulation
relative efficiency
