scientific article; zbMATH DE number 3997027
From MaRDI portal
Publication:4723167
[For the entire collection see Zbl 0611.00024.] Some of the factors to be considered when applying the techniques of numerical linear algebra to statistical problems are discussed with reference to three particular examples: the use of the normal equations in regression problems; the use of perturbation theory to assess the effects of errors in regression matrices; and the phenomenon of benign degeneracy, in which the numerical problem becomes more difficult even when the associated statistical problem becomes easier.
Recommendations
Cited in
(8)- Error bounds for computed least squares estimators
- Centering and scaling of regression algorithms in the face of iii–conditioning
- The Early Use of Matrix Diagonal Increments in Statistical Problems
- scientific article; zbMATH DE number 741203 (Why is no real title available?)
- scientific article; zbMATH DE number 1058076 (Why is no real title available?)
- Exploiting lower precision arithmetic in solving symmetric positive definite linear systems and least squares problems
- Sparse linear least-squares problems
- All possible subset regressions using the QR decomposition
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4723167)