Block imbedding and interlacing results for normal matrices
نویسندگان
چکیده
A pair of matrices is said to be imbeddable precisely when one is an isometric projection of the other on a suitable subspace. The concept of imbedding has been the subject of extensive study. Particular emphasis has been placed on relating the spectra of the matrices involved, especially when both matrices are Hermitian or normal. In this paper, the notion of block imbedding is introduced and shown to be intimately connected to an extension of interlacing for eigenvalues of normal matrices. Thus, a generalization of a classic Theorem of K. Fan and G. Pall is obtained, which is then applied to yield bounds on the number of eigenvalues of a block imbeddable pair in a closed, convex set. Moreover, a wide class of normal matrices, for which block imbedding applies, is indicated. Finally, comments and links on the necessary imbedding conditions of D. Carlson and E.M. de Sa, and J.P. Queiro and A.L. Duarte are provided.
منابع مشابه
Interlacing Eigenvalues and Graphs
We give several old and some new applications of eigenvalue interlacing to matrices associated to graphs. Bounds are obtained for characteristic numbers of graphs, such as the size of a maximal (colclique, the chromatic number, the diameter, and the bandwidth, in terms of the eigenvalues of the Standard adjacency matrix or the Laplacian matrix. We also deal with inequalities and regularity resu...
متن کاملAn interlacing property of eigenvalues strictly totally positive matrices
We prove results concerning the interlacing of eigenvalues of principal submatrices of strictly totally positive (STP) matrices.
متن کاملAn Interlacing Property of Eigenvalues of Strictly Totally Positive Matrices
We prove results concerning the interlacing of eigenvalues of principal submatrices of strictly totally positive matrices. §
متن کاملSelf-interlacing polynomials II: Matrices with self-interlacing spectrum
An n×n matrix is said to have a self-interlacing spectrum if its eigenvalues λk , k = 1, . . . ,n, are distributed as follows: λ1 >−λ2 > λ3 > · · ·> (−1)λn > 0. A method for constructing sign definite matrices with self-interlacing spectrum from totally nonnegative ones is presented. This method is applied to bidiagonal and tridiagonal matrices. In particular, a result by O. Holtz on the spectr...
متن کاملInterlacing Properties for Hermitian Matrices Whose Graph is a Given Tree
We extend some interlacing properties of the eigenvalues of tridiagonal matrices to Hermitian matrices whose graph is a tree. We also give a graphical interpretation of the results. We use the work on matchings polynomials by O.L. Heilmann and E.H. Lieb.
متن کامل