Eigenvalue computation for unitary rank structured matrices

In this paper we describe how to compute the eigenvalues of a unitary rank structured matrix in two steps. First we perform a reduction of the given matrix into Hessenberg form, next we compute the eigenvalues of this resulting Hessenberg matrix via an implicit QR-algorithm. Along the way, we explain how the knowledge of a certain ‘shift’ correction term to the structure can be used to speed up the QR-algorithm for unitary Hessenberg matrices, and how this observation was implicitly used in a paper due to William B. Gragg. We also treat an analogue of this observation in the Hermitian tridiagonal case.     

Matrices with prescribed Ritz values

On the way to establishing a commutative analog to the Gelfand-Kirillov theorem in Lie theory, Kostant and Wallach produced a decomposition of M(n) which we will describe in the language of linear algebra. The “Ritz values” of a matrix are the eigenvalues of its leading principal submatrices of order m=1,2,…,n. There is a unique unit upper Hessenberg matrix H with those eigenvalues. For real symmetric matrices with interlacing Ritz values, we extend their analysis to allow eigenvalues at successive levels to be equal. We also decide whether given Ritz values can come from a tridiagonal matrix.     

Functions of a matrix and Krylov matrices

For a given nonderogatory matrix A, formulas are given for functions of A in terms of Krylov matrices of A. Relations between the coefficients of a polynomial of A and the generating vector of a Krylov matrix of A are provided. With the formulas, linear transformations between Krylov matrices and functions of A are introduced, and associated algebraic properties are derived. Hessenberg reduction forms are revisited equipped with appropriate inner products and related properties and matrix factorizations are given.     

Moving zeros among matrices

We investigate the zero-patterns that can be created by unitary similarity in a given matrix, and the zero-patterns that can be created by simultaneous unitary similarity in a given sequence of matrices. The latter framework allows a “simultaneous Hessenberg” formulation of Pati’s tridiagonal result for 4 × 4 matrices. This formulation appears to be a strengthening of Pati’s theorem. Our work depends at several points on the simplified proof of Pati’s result by Davidson and Djokovi?. The Hessenberg approach allows us to work with ordinary similarity and suggests an extension from the complex to arbitrary algebraically closed fields. This extension is achieved and related results for 5 × 5 and larger matrices are formulated and proved.     

On the Inverses of Brownian and Brownian-Like Matrices

In several papers by F. Valvi, sufficient conditions are given for Brownian and Brownian-like matrices to have Hessenberg inverses. We interpret these conditions from the viewpoint of familiar facts related to matrices of small triangular rank. This allows us to formulate more general assertions on the Hessenberg property of the inverse. Moreover, we explicitly find the structure of the inverse of a Brownian matrix under a certain natural “irreducibility” condition. This structure is similar to the well-known structure of the inverse of an irreducible tridiagonal matrix. Furthermore, we show that the parameters defining the inverse of an n X n Brownian matrix can be calculated in O(n) arithmetic operations.     

The inverses of Toeplitz band matrices

The elements of the inverse of a Toeplitz band matrix are given in terms ofthe solution of a difference equation. The expression for these elements is a quotient of determinants whose orders depend the number of nonzero superdiagonals but not on the order of the matrix. Thus, the formulae are particularly simple for lower triangular and lower Hessenberg Toeplitz matrices. When the number of nonzero superdiagonals is small, sufficient conditions on the solution of the abovementioned difference equation can be given to ensure that the inverse matrix is positive. If the inverse is positive, the row sums can be expressed in terms of the solution of the difference equation.     

The Drazin Inverse of Hessenberg Matrices

The Drazin inverse of a lower Hessenberg matrix is considered. If $A$ is a singular lower Hessenberg matrix and $a_{i,i+1}=\neq 0,i=1,2,\cdots,n-1$, then $A^D$ can be given, and expressed explicitly by elements of $A$. The structure of the Drazin inverse of a lower Hessenberg matrix is also studied.     

An algorithm for determining if the inverse of a strictly diagonally dominant Stieltjes matrix is strictly ultrametric

Summary It was recently shown that the inverse of a strictly ultrametric matrix is a strictly diagonally dominant Stieltjes matrix. On the other hand, as it is well-known that the inverse of a strictly diagonally dominant Stieltjes matrix is a real symmetric matrix with nonnegative entries, it is natural to ask, conversely, if every strictly diagonally dominant Stieltjes matrix has a strictly ultrametric inverse. Examples show, however, that the converse is not true in general, i.e., there are strictly diagonally dominant Stieltjes matrices in ^n×n (for everyn3) whose inverses are not strictly ultrametric matrices. Then, the question naturally arises if one can determine which strictly diagonally dominant Stieltjes matrices, in ^n×n (n3), have inverses which are strictly ultrametric. Here, we develop an algorithm, based on graph theory, which determines if a given strictly diagonally dominant Stieltjes matrixA has a strictly ultrametric inverse, where the algorithm is applied toA and requires no computation of inverse. Moreover, if this given strictly diagonally dominant Stieltjes matrix has a strictly ultrametric inverse, our algorithm uniquely determines this inverse as a special sum of rank-one matrices.Research supported by the National Science FoundationResearch supported by the Deutsche Forschungsgemeinschaft     

Thin Hessenberg pairs

A square matrix is called Hessenberg whenever each entry below the subdiagonal is zero and each entry on the subdiagonal is nonzero. Let V denote a nonzero finite-dimensional vector space over a field K. We consider an ordered pair of linear transformations A:V→V and A^∗:V→V which satisfy both (i) and (ii) below.

(i)

There exists a basis for V with respect to which the matrix representing A is Hessenberg and the matrix representing A^∗ is diagonal.

(ii)

There exists a basis for V with respect to which the matrix representing A is diagonal and the matrix representing A^∗ is Hessenberg.

We call such a pair a thin Hessenberg pair (or TH pair). This is a special case of a Hessenberg pair which was introduced by the author in an earlier paper. We investigate several bases for V with respect to which the matrices representing A and A^∗ are attractive. We display these matrices along with the transition matrices relating the bases. We introduce an “oriented” version of called a TH system. We classify the TH systems up to isomorphism.     

Orthogonal polynomials and measures on the unit circle. The Geronimus transformations

In this paper we analyze a perturbation of a nontrivial positive measure supported on the unit circle. This perturbation is the inverse of the Christoffel transformation and is called the Geronimus transformation. We study the corresponding sequences of monic orthogonal polynomials as well as the connection between the associated Hessenberg matrices. Finally, we show an example of this kind of transformation.     

Green’s matrices

Our goal is to identify and understand matrices A that share essential properties of the unitary Hessenberg matrices M that are fundamental for Szegö’s orthogonal polynomials. Those properties include: (i) Recurrence relations connect characteristic polynomials {r_k(x)} of principal minors of A. (ii) A is determined by generators (parameters generalizing reflection coefficients of unitary Hessenberg theory). (iii) Polynomials {r_k(x)} correspond not only to A but also to a certain “CMV-like” five-diagonal matrix. (iv) The five-diagonal matrix factors into a product BC of block diagonal matrices with 2 × 2 blocks. (v) Submatrices above and below the main diagonal of A have rank 1. (vi) A is a multiplication operator in the appropriate basis of Laurent polynomials. (vii) Eigenvectors of A can be expressed in terms of those polynomials.Conditions (v) connects our analysis to the study of quasi-separable matrices. But the factorization requirement (iv) narrows it to the subclass of “Green’s matrices” that share Properties (i)-(vii).The key tool is “twist transformations” that provide 2ⁿ matrices all sharing characteristic polynomials of principal minors with A. One such twist transformation connects unitary Hessenberg to CMV. Another twist transformation explains findings of Fiedler who noticed that companion matrices give examples outside the unitary Hessenberg framework. We mention briefly the further example of a Daubechies wavelet matrix. Infinite matrices are included.     

On zero-pattern invariant properties of structured matrices

It is interesting that inverse M-matrices are zero-pattern (power) invariant. The main contribution of the present work is that we characterize some structured matrices that are zero-pattern (power) invariant. Consequently, we provide necessary and sufficient conditions for these structured matrices to be inverse M-matrices. In particular, to check if a given circulant or symmetric Toeplitz matrix is an inverse M-matrix, we only need to consider its pattern structure and verify that one of its principal submatrices is an inverse M-matrix.     

A new algorithm for computing the inverse and the determinant of a Hessenberg matrix

In this paper, a new recursive symbolic computational Hessenberg matrix algorithm is presented to compute the inverse and the determinant of a Hessenberg matrix.     

A Jacobi matrix inverse eigenvalue problem with mixed data

In this paper, the inverse eigenvalue problem of reconstructing a Jacobi matrix from part of its eigenvalues and its leading principal submatrix is considered. The necessary and sufficient conditions for the existence and uniqueness of the solution are derived. Furthermore, a numerical algorithm and some numerical examples are given.     

The rational Krylov algorithm for nonsymmetric eigenvalue problems. III: Complex shifts for real matrices

A new algorithm for the computation of eigenvalues of a nonsymmetric matrix pencil is described. It is a generalization of the shifted and inverted Lanczos (or Arnoldi) algorithm, in which several shifts are used in one run. It computes an orthogonal basis and a small Hessenberg pencil. The eigensolution of the Hessenberg pencil, gives Ritz approximations to the solution of the original pencil. It is shown how complex shifts can be used to compute a real block Hessenberg pencil to a real matrix pair.Two applicationx, one coming from an aircraft stability problem and the other from a hydrodynamic bifurcation, have been tested and results are reported.Dedicated to Carl-Erik Fröberg on the occasion of his 75th birthday.     

On inverses of Hessenberg matrices

The lower half of the inverse of a lower Hessenberg matrix is shown to have a simple structure. The result is applied to find an algorithm for finding the inverse of a tridiagonal matrix. With minor modifications, the technique applies to block Hessenberg matrices.     

Principal minors, Part II: The principal minor assignment problem

The inverse problem of finding a matrix with prescribed principal minors is considered. A condition that implies a constructive algorithm for solving this problem will always succeed is presented. The algorithm is based on reconstructing matrices from their principal submatrices and Schur complements in a recursive manner. Consequences regarding the overdeterminancy of this inverse problem are examined, leading to a faster (polynomial time) version of the algorithmic construction. Care is given in the MATLAB^® implementation of the algorithms regarding numerical stability and accuracy.     

Newton Generalized Hessenberg method for solving nonlinear systems of equations

In this paper, we give and analyze a Finite Difference version of the Generalized Hessenberg (FDGH) method. The obtained results show that applying this method in solving a linear system is equivalent to applying the Generalized Hessenberg method to a perturbed system. The finite difference version of the Generalized Hessenberg method is used in the context of solving nonlinear systems of equations using an inexact Newton method. The local convergence of the finite difference versions of the Newton Generalized Hessenberg method is studied. We obtain theoretical results that generalize those obtained for Newton-Arnoldi and Newton-GMRES methods. Numerical examples are given in order to compare the performances of the finite difference versions of the Newton-GMRES and Newton-CMRH methods. This revised version was published online in June 2006 with corrections to the Cover Date.     

Mixed-type reverse order law for products of three operators

We present new results related to the mixed-type reverse order law for the Moore-Penrose inverse of various products of three operators on Hilbert spaces. Some finite dimensional results are extended to infinite dimensional settings.     

The centro-invertible matrix: A new type of matrix arising in pseudo-random number generation

This paper defines a new type of matrix (which will be called a centro-invertible matrix) with the property that the inverse can be found by simply rotating all the elements of the matrix through 180 degrees about the mid-point of the matrix. Centro-invertible matrices have been demonstrated in a previous paper to arise in the analysis of a particular algorithm used for the generation of uniformly-distributed pseudo-random numbers.An involutory matrix is one for which the square of the matrix is equal to the identity. It is shown that there is a one-to-one correspondence between the centro-invertible matrices and the involutory matrices. When working in modular arithmetic this result allows all possible k by k centro-invertible matrices with integer entries modulo M to be enumerated by drawing on existing theoretical results for involutory matrices.Consider the k by k matrices over the integers modulo M. If M takes any specified finite integer value greater than or equal to two then there are only a finite number of such matrices and it is valid to consider the likelihood of such a matrix arising by chance. It is possible to derive both exact expressions and order-of-magnitude estimates for the number of k by k centro-invertible matrices that exist over the integers modulo M. It is shown that order (√N) of the N=M^(k2) different k by k matrices modulo M are centro-invertible, so that the proportion of these matrices that are centro-invertible is order (1/√N).