节点文献
半序约束下多维正态总体均值和协方差阵的最大似然估计
Maximum Likelihood Estimation of Mean Vectors and Covariance Matrices from Multivariate Normal Distribution under Simultaneous Order Restrictions
【摘要】 对给定K个P维正态总体,未知均值和协方差阵分别为θi和Λi,i=1,2,…,k,本文考虑均值和协方差阵之间都在一个简单半序约束θ1≤θ2≤…≤θk,Λ1≥Λ2≥…≥Λk>0条件下的估计问题.讨论θi和Λi的最大似然估计的性质,并给出一个求解的迭代方法.
【Abstract】 For kp-multiple normal populations with unknown mean vector θi and unknown covariance matrix Λi,i=1,2,…,k,assume that there are some order restrictions among the mean vectors and covariance matrices,respectively,for example,simple order restrictions:θ1≤θ2≤…≤θk and Λ1≥Λ2≥…≥Λk>0.Some properties of maximum likelihood estimations of θ′is and Λ′is are discussed and an algorithm of obstaining the maximum likelihood estimators under the order restriction is proposed.
【关键词】 最大似然估计;
保序回归;
迭代算法;
【Key words】 Maximum likelihood estimation; Isotonic regression; Iterative algorithm;
【Key words】 Maximum likelihood estimation; Isotonic regression; Iterative algorithm;
【基金】 辽宁省自然科学基金赞助项目(20031066)
- 【文献出处】 应用数学 ,Mathematica Applicata , 编辑部邮箱 ,2005年03期
- 【分类号】O212.1
- 【被引频次】2
- 【下载频次】157