摘要
在不确定性线性模型中,参数通常带有不确定的先验信息,它们影响着参数估计及其精度.本文利用椭球集合对含有先验信息的参数进行描述,建立基于椭球不确定性的线性模型,并提出相应的估计算法.本文主要采用直接迭代算法解决参数估计中的权值分配问题,给出参数估计可容许性的充要条件.最后,通过实证分析算法的有效性,解决权值分配对平差解的影响.
In uncertain linear models,parameters usually carry uncertain prior information,which affects parameter estimation and accuracy.In this paper,by using a set of ellipsoids to describe the parameters containing prior information,a linear model based on the uncertainty of the ellipsoid is established,and a corresponding estimation algorithm is proposed.We mainly adopted the direct iterative algorithm to solve the problem of weight allocation in parameter estimation,and put forward sufficient and necessary conditions for the admissibility of parameter estimation.Finally,the effectiveness of the algorithm is analyzed empirically,and the influence of weight distribution on the adjustment solution is solved.
作者
王志忠
邹航
Wang Zhizhong;Zou Hang(School of Mathematics and Statistics,Central South University Changsha,Hunan 410083,China)
出处
《数学理论与应用》
2020年第2期118-128,共11页
Mathematical Theory and Applications
基金
国家自然科学基金重点项目:41730105
关键词
不确定性
回归模型
椭球约束
容许估计
迭代算法
Uncertainty
Regression model
Ellipsoid constraint
Allowable estimation
Iterative algorithm
