Robust estimation for correlated observations: two local sensitivity-based downweighting strategies |
| |
| Authors: | Jianfeng Guo Jikun Ou Haitao Wang |
| |
| Institution: | (1) Present address: Engineering Geodesy and Measurement Systems, Graz University of Technology, Steyrergasse 30, 8010 Graz, Austria;(2) Department of Geomatics Engineering, University of Calgary, 2500 University Drive NW, Calgary, AB, T2N 1N4, Canada |
| |
| Abstract: | An iterative robust estimation procedure for correlated observations is proposed, in which the a-prior correlation coefficient
matrix is not updated to alleviate the computational burden. Selection of the downweighting strategy plays a key role in the
proposed method. Two local sensitivity-based strategies, one is based on the uniformly most powerful test statistics, the
other is based on the standardized least squares residuals, are developed and analyzed. Monte Carlo simulations in the GPS
network adjustment scenario demonstrate that, the two strategies can provide a certain resistance against the deteriorating
effect of outlying observations on the parameter estimates; the former downweighting strategy is superior to the latter one,
both in terms of robustness and computational efficiency. |
| |
| Keywords: | |
| 本文献已被 SpringerLink 等数据库收录！ |
|
