| MSE准则下岭-主成分组合估计与LS估计的比较与选择 |
| |
| 引用本文： | 归庆明,韩松辉,宫轶松,姚绍文,李国重.MSE准则下岭-主成分组合估计与LS估计的比较与选择[J].大地测量与地球动力学,2005,25(4):79-83. |
| |
| 作者姓名： | 归庆明 韩松辉 宫轶松 姚绍文 李国重 |
| |
| 作者单位： | 信息工程大学理学院,郑州,450001;信息工程大学测绘学院,郑州,450052;信息工程大学理学院,郑州,450001 |
| |
| 基金项目： | 国家重点基础研究发展计划(973计划);中国科学院资助项目;山东省重点实验室基金;河南省自然科学基金 |
| |
| 摘 要： | 研究了测量平差Gauss—Markov模型中岭-主成分组合估计与LS估计的比较与选择问题。首先在均方误差（MSE）准则下对岭-主成分组合估计与LS估计进行了比较，得到了岭-主成分组合估计优于LS估计的椭球条件；然后运用Monte Carlo方法对这些条件进行了假设检验；最后通过数值实验说明，在一定显著性水平下当原假设被接受时，可采用岭-主成分组合估计对LS估计做出比较有效的改进，当原假设被拒绝时，应该仍采用LS估计。
|
| 关 键 词： | 岭-主成分组合估计 LS估计 均方误差准则(MSE) Monte Carlo方法 假设检验 |
| 文章编号： | 1671-5942(2005)04-0079-05 |
| 修稿时间： | 2005年2月19日 |
COMPARISON AND SELECTION BETWEEN CRPC ESTIMATOR AND LS ESTIMATOR UNDER MSE CRITERION |
| |
| Gui Qingming,Han Songhui,Gong Yisong,Yao Shaowen,Li Guozhong.COMPARISON AND SELECTION BETWEEN CRPC ESTIMATOR AND LS ESTIMATOR UNDER MSE CRITERION[J].Journal of Geodesy and Geodynamics,2005,25(4):79-83. |
| |
| Authors: | Gui Qingming Han Songhui Gong Yisong Yao Shaowen Li Guozhong |
| |
| Abstract: | The problem of selection between combining ridge and principal component (CRPC) estimator and LS estimator in Gauss-Markov model is studied. Firstly, the comparison between CRPC estimator and LS estimator is conducted by use of the criterion of mean squared error, and the ellipsoid condition showing the superiority of CRPC estimator over the LS estimator has been obtained. Then, the conditions are tested hypotectically by use of Monte Carlo simulation. Finally, the computational results demonstrate that if the null hypothesis is accepted with a significance level, using the CRPC estimator will improve LS estimator more effectively. On the contrary, if the null hypothesis is rejected,the LS estimator is still a good way. |
| |
| Keywords: | combining ridge and principal component estimator(CRPC) LS estimator mean squared error Monte Carlo method hypothesis test |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
