Invariance of numerical character of matrix products and their statistical applications (Q2764278)
From MaRDI portal
 | This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Invariance of numerical character of matrix products and their statistical applications |
scientific article; zbMATH DE number 1690300
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Invariance of numerical character of matrix products and their statistical applications |
scientific article; zbMATH DE number 1690300 |
Statements
10 September 2002
0 references
\(g\)-inverse
0 references
invariance of matrix products
0 references
matrix rank
0 references
Markov processes
0 references
least squares estimator
0 references
Gauss-Markov model
0 references
0.89162576
0 references
0.8897255
0 references
0.8850476
0 references
0.8776506
0 references
0.87638515
0 references
0.8718627
0 references
0.8706078
0 references
0.86950666
0 references
Invariance of numerical character of matrix products and their statistical applications (English)
0 references
Necessary and sufficient conditions are derived for the product \(AB^-C\) to be invariant with respect to the minimum norm \(g\)-inverse \(B^-\). The relationships among the ordinary least squares estimator, weight least squares estimator, and the best least squares estimator are investigated under the general Gauss-Markov model.
0 references
