Data perturbation analysis of a linear model①
(1. Geoscience Deptartment, Shandong University of Science and Technology,
Taian 271019, P.R.China;
2. Positioning Metrology and Surveying Group, European Organization for Nuclear Research (CERN),Geneva, Switzerland)
Taian 271019, P.R.China;
2. Positioning Metrology and Surveying Group, European Organization for Nuclear Research (CERN),Geneva, Switzerland)
Abstract: The linear model features were carefully studied in the cases of data perturbation and mean shift perturbation. Some important features were also proved mathematically. The results show that the mean shift perturbation is equivalent to the data perturbation, that is, adding a parameter to an observation equation means that this set of data is deleted from the data set. The estimate of this parameter is its predicted residual in fact.
Key words: linear model; data perturbation; mean shift perturbation