scientific article; zbMATH DE number 1389757
Estimates for the eigenvalues are given when the off-diagonal elements in a tridiagonal matrix are replaced by zero. The results can be applied to \(QR\) or \(QL\) algorithm with Rayleigh quotient shift or Wilkinson shift. A quadratic estimate in general is presented. The difference between the eigenvalues of an unreduced symmetric tridiagonal matrix \(T\) and an approximate matrix \[ \widehat T=\left[\begin{smallmatrix} \alpha_1 & \\ & T_{2,n}\end{smallmatrix}\right] \] are deduced and illustrated by two examples. The estimate improves the Hager's result of \textit{W. W. Hager} and \textit{R. N. Pederson} [Linear Algebra Appl. 42, 39-55 (1982; Zbl 0476.15008)]. The estimate in another case for a matrix of the type \[ \widehat T=\left[\begin{smallmatrix} T_{1, i-1} & &\\ & \alpha_i &\\ & & T_{i+1,n}\end{smallmatrix}\right] \] is discussed and an example is given.
- \(QR\)-like algorithms for eigenvalue problems
- Perturbation in eigenvalues of a symmetric tridiagonal matrix
- scientific article; zbMATH DE number 3913545 (Why is no real title available?)
- scientific article; zbMATH DE number 727777 (Why is no real title available?)
- Matrix shapes invariant under the symmetric QR algorithm
- A symmetric structure-preserving {\(\Gamma\)}QR algorithm for linear response eigenvalue problems
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 Q4934210)