首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于Neumann级数的有偏估计方法
引用本文:杨秋伟,白志超,李翠红.基于Neumann级数的有偏估计方法[J].大地测量与地球动力学,2020,40(5):512-516.
作者姓名:杨秋伟  白志超  李翠红
作者单位:绍兴文理学院土木工程学院
基金项目:国家自然科学基金(11202138)。
摘    要:针对病态最小二乘问题,提出基于Neumann级数的有偏估计方法。由该方法建立的有偏估计公式,包含现有的最小二乘估计、岭估计和广义岭估计公式,建立有偏估计和无偏估计的本质联系,并可以据此引申出一系列新的有偏估计形式。在这一系列的有偏估计中,存在着比岭估计或广义岭估计更接近于真值的解。以一个病态方程组为例对该方法进行验证表明,当观测向量中含有误差时,由最小二乘估计所得结果和真值误差较大,而各级有偏估计的结果和真值更加接近,且二级和三级有偏估计结果比相应的岭估计更接近于真值。

关 键 词:病态方程  Neumann级数  最小二乘估计  岭估计  

Biased Estimation Method Based on Neumann Series
YANG Qiuwei,BAI Zhichao,LI Cuihong.Biased Estimation Method Based on Neumann Series[J].Journal of Geodesy and Geodynamics,2020,40(5):512-516.
Authors:YANG Qiuwei  BAI Zhichao  LI Cuihong
Institution:(Department of Civil Engineering,Shaoxing University,508 West-Huancheng Road,Shaoxing 312000,China)
Abstract:In this paper, we propose a biased estimation method based on Neumann series to solve the ill-conditioned least squares problem. The biased estimation formulas established by the proposed method can include the existing least squares estimation, ridge estimation and generalized ridge estimation formulas. Using the proposed method, we can establish the essential relationship between biased estimation and unbiased estimation. In addition, we can derive a series of new biased estimations from the proposed method. In the series of biased estimates, there are solutions that are closer to the true value than ridge estimates or generalized ridge estimates. An ill-conditioned system of equations is taken as an example to verify the proposed method. The results show that when the observation vectors contain noise, the errors between the results obtained by least squares estimation and the true values are very large. However, the results of biased estimates at all levels are closer to the true values. Moreover, the results of biased estimates at two and three levels are closer to the true values than the corresponding ridge estimates. It is shown that the method proposed in this paper provides a new way to solve the ill conditioned least squares problem.
Keywords:ill-conditioned equation  Neumann series  least squares estimate  ridge estimate  
本文献已被 CNKI 维普 等数据库收录!
点击此处可从《大地测量与地球动力学》浏览原始摘要信息
点击此处可从《大地测量与地球动力学》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号