摘要
本文从有偏估计类中的岭型组合主成分估计出发 ,结合抗差估计理论 ,利用抗差M估计模型 ,提出了一种新的抗差有偏估计法———抗差岭型组合主成分估计。推导了平差参数的抗差岭型组合主成分估计解 ,以及平差参数的验后精度和误差影响函数。计算结果表明 ,抗差岭型组合主成分估计不但能克服法方程系数阵病态性的影响 ,而且能有效地抵制观测值中精差的异常干扰 。
In this paper,the ridge combined the principal components estimation of adjust parameter is united with the robust estimation,using the model of robust M estimation and building the corresponding model of the robust ridge combined principal components estimation of adjust parameter. The paper gives a test in the end,Test results show that the robust ridge combined principal components estimation of adjust parameter not only can improve the ill condition of normal equations,but also resists the effect of course errors effectively.
出处
《测绘科学》
CSCD
2004年第2期50-52,共3页
Science of Surveying and Mapping
关键词
岭估计
主成分估计
抗差估计
影响函数
ridge estimation
principal components estimation
robust estimation
influence function
