On the joint robust estimation of parameters and variance |
| |
Authors: | Jia Pei-zhang |
| |
Affiliation: | Institute of System Science, Chinese Academy of Sciences, Beijing 100080, China |
| |
Abstract: | A method of joint estimation of the parameters and the variance σ2 in the linear model is presented in this paper. The M-estimators with finite rejection points are adopted for estimating the parameters. In order to estimate the variance σ2, the Grubbs statistics and the Kurtosis test statistics are adopted to test the residual sequence {;ri};. And the sample variance of {;ri}; after discarding the outliers is taken as the estimation of σ2. This method of estimating σ2 is less computational demanding and more accurate, compared with the well-known method which takes 1.483 medi |ri| as the estimation of σ. The breakdown point of the above estimation of σ2 is more than 20%, while the brekdown point is less than 10%, if only the Grubbs statistics is used. |
| |
Keywords: | parameters estimation variance estimate |
本文献已被 ScienceDirect 等数据库收录! |
|