
Open access
Date
2011Type
- Report
ETH Bibliography
yes
Altmetrics
Abstract
In this paper, we analyze the distribution of the eigenvalues of glued tridiagonal matrices. Such matrices provide a useful class of test matrix because, despite being unreduced, a glued matrix can have some eigenvalues agreeing to hundreds of decimal places. A glued matrix can be obtained from a direct sum of p copies of an unreduced symmetric tridiagonal matrix T by modifying the junctions, in one of two ways, so that the new matrix has no zero off-diagonal entries. We exhibit, in gradually increasing detail, how width and placement of the eigenvalue clusters of a glued matrix depend on T, on p, and on the strength of the glue γ. Show more
Permanent link
https://doi.org/10.3929/ethz-a-006810497Publication status
publishedJournal / series
Technical Report / ETH Zurich, Department of Computer ScienceVolume
Publisher
ETH, Department of Computer ScienceSubject
Rank-1 gluing; Rank-2 gluing; EIGENVALUES OF MATRICES (ALGEBRA); Eigenvalue clusters; MATRIZEN UND LINEARE ABBILDUNGEN (ALGEBRA); Glued matrix; MATRIZENEIGENWERTE (ALGEBRA); MATRICES AND LINEAR MAPPINGS (ALGEBRA)Organisational unit
02150 - Dep. Informatik / Dep. of Computer Science
Notes
Technical Reports D-INFK.More
Show all metadata
ETH Bibliography
yes
Altmetrics