scientific article; zbMATH DE number 17452
[For the entire collection see Zbl 0728.00018.] This paper examines the factors influencing the accuracy of the solution when a symmetric tridiagonal matrix eigenproblem is solved. The bisection method and inverse iteration are used. Bisection produces eigenvalues to high relative accuracy. Inverse iteration results in eigenvectors with quality dependent on the spacing of the eigenvalue spectrum and on the starting vector. The effects on the accuracy of the solution of using random starting vectors are examined. In addition, some uses of statistical analysis in the design of an inverse iteration algorithm are presented.
- Increasing efficiency of inverse iteration
- Practical methods for evaluating the accuracy of the eigenelements of a symmetric matrix
- Current inverse iteration software can fail
- scientific article; zbMATH DE number 4074343 (Why is no real title available?)
- Improving the Accuracy of Inverse Iteration
- Computing an Eigenvector with Inverse Iteration
- scientific article; zbMATH DE number 1910736 (Why is no real title available?)
- The Godunov-inverse iteration: A fast and accurate solution to the symmetric tridiagonal eigenvalue problem
- A statistic on involutions
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 Q3974772)
