Transforming an error-tolerant separable matrix to an error-tolerant disjunct matrix

Recently, Chen and Hwang [H.B. Chen, F.K. Hwang, Exploring the missing link among d-separable, -separable and d-disjunct matrices, Discrete Applied Mathematics 133 (2007) 662-664] provided a method for transforming a separable matrix to a disjunct matrix. In [D.Z. Du, F.K. Hwang, Pooling Designs and Nonadaptive Group Testing — Important Tools for DNA Sequencing, World Scientific, 2006], Du and Hwang attempted to extend this result to its error-tolerant version; unfortunately, they gave an incorrect extension. This note gives a solution to this problem.     

Improved estimation of a covariance matrix in an elliptically contoured matrix distribution

In this paper, the problem of estimating the covariance matrix of the elliptically contoured distribution (ECD) is considered. A new class of estimators which shrink the eigenvalues towards their arithmetic mean is proposed. It is shown that this new estimator dominates the unbiased estimator under the squared error loss function. Two special classes of ECD, namely, the multivariate-elliptical t distribution and the ε-contaminated normal distribution are considered. A simulation study is carried out and indicates that this new shrinkage estimator provides a substantial improvement in risk under most situations.     

On schur complements in an ep matrix

It is established that under certain conditions a Schur complement in an EP matrix is as well an EP matrix. As an application a decomposition of a partitioned matrix into a sum of EP matrices is given.     

On an interval-arithmetic matrix method

Hansen and Smith have proposed a method for solving linear algebraic systems with interval coefficients which produces good results if the intervals are narrow. In this paper we show that the error of their method isO(W ²), whereW is the width of the set of coefficients. The effects of round-off errors are not considered.Supported in part by NSF grant GJ-797.     

LDU decomposition of an extension matrix of the Pascal matrix

Let Ω_m,n(α,β,γ) denote a set of all elements of weighted lattice paths with weight (α,β,γ) in the xy-plane from (0,0) to (m,n) such that a vertical step V=(0,1), a horizontal step H=(1,0), and a diagonal step D=(1,1) are endowed with weights α,β, and γ respectively and let ω(Ω_m,n(α,β,γ)) denote the weight of Ω_m,n(α,β,γ) defined by

     

Computing an eigenvector of an inverse Monge matrix in max-plus algebra

The problem of computing an eigenvector of an inverse Monge matrix in max-plus algebra is addressed. For a general matrix, the problem can be solved in at most O(n³) time. This note presents an O(n²) algorithm for computing one max-plus algebraic eigenvector of an inverse Monge matrix . It is assumed that is irreducible.     

Hermitian semidefinite matrix means and related matrix inequalities—an introduction

A survey of matrix means and matrix inequalities is presented with the work of Ando, Anderson and Duffin being showcased. The classical Arithmetic-Geometric-Harmonic Mean for two Hermitian positive semidefinite matrices is given. Other classical means such as the Gaussian Mean, power means and the symmetric function means are also discussed.     

The scattering matrix with respect to an Hermitian matrix of a graph

Recently, Gnutzmann and Smilansky [5] presented a formula for the bond scattering matrix of a graph with respect to an Hermitian matrix. We present another proof for this formula by a technique use in the zeta function of a graph. Furthermore, we generalize Gnutzmann and Smilansky's formula to a regular covering of a graph. Finally, we define an L-function of a graph, and present a determinant expression. As a corollary, we express the generalization of Gnutzmann and Smilansky's formula to a regular covering of a graph by using its L-functions.     

Eigenvalue bounds for an alignment matrix in manifold learning

This paper presents a spectral analysis for an alignment matrix that arises in reconstruction of a global coordinate system from local coordinate systems through alignment in manifold learning. Some characterizations of its eigenvalues and its null space as well as a lower bound for the smallest positive eigenvalue are given, which generalize earlier results of Li et al. (2007) [4] to include a more general situation that arises in alignments of local sections of different dimensions. Our results provide a theoretical understanding of the Local Tangent Space Alignment (LTSA) method (Zhang and Zha (2004) [12]) for nonlinear manifold learning and address some computational issues related to the method.     

Implementation of sparse matrix algorithms in an advection-diffusion-chemistry module

A two-dimensional advection-diffusion-chemistry module of a large-scale environmental model is taken. The module is described mathematically by a system of partial differential equations. Sequential splitting is used in the numerical treatment. The non-linear chemistry is most time-consuming part and it is handled by six implicit algorithms for solving ordinary differential equations. This leads to the solution of very long sequences of systems of linear algebraic equations. It is crucial to solve these systems efficiently. This is achieved by applying four different algorithms. The numerical results indicate that the algorithms based on a preconditioned sparse matrix technique and on a specially designed algorithm for the particular problem under consideration perform best.     

Almost orthogonal submatrices of an orthogonal matrix

Lett≥1 and letn, M be natural numbers,n<M. Leta=(a _i,j ) be ann xM matrix whose rows are orthonormal. Suppose that the ℓ₂-norms of the columns ofA are uniformly bounded. Namely, for allj Using majorizing measure estimates we prove that for every ε>0 there exists, a setI ⊃ {1,…,M} of cardinality at most such that the matrix , whereA _I =(a _i,j ) _j∈I , acts as a (1+ε)-isomorphism from ℓ ₂ ⁿ into . Research supported in part by a grant of the US-Israel BSF. Part of this research was performed when the author held a postdoctoral position at MSRI. Research at MSRI was supported in part by NSF grant DMS-9022140.     

Anymatrix: an extensible MATLAB matrix collection

Numerical Algorithms - Anymatrix is a MATLAB toolbox that provides an extensible collection of matrices with the ability to search the collection by matrix properties. Each matrix is implemented as...     

Noncommutative symmetric functions and an amazing matrix

We present a simple way to derive the results of Diaconis and Fulman [P. Diaconis, J. Fulman, Foulkes characters, Eulerian idempotents, and an amazing matrix, arXiv:1102.5159] in terms of noncommutative symmetric functions.     

Normal equivalent to an arbitrary diagonalizable matrix

The aim of this paper is to reduce the eigenvalue problem of a diagonalizable matrix to the eigenvalue problem of an equivalent normal matrix. We use for this purpose a minimization strategy, which is also applicable for transforming an arbitrary nondiagonalizable matrix to an almost normal one.     

Sensitivity in data envelopment analysis using an approximate inverse matrix

In this paper, sensitivity analysis of the Charnes–Cooper–Rhodes model in data envelopment analysis (DEA) is studied for the case of perturbation of all outputs and of all inputs of an efficient decision-making unit (DMU). Using an approximate inverse of the perturbed optimal basis matrix, an approximate preservation of efficiency for an efficient DMU under these perturbations is considered. Sufficient conditions for an efficient DMU to preserve its efficiency are obtained in that case. An illustrative example is provided.