p.n.p.矩阵的一些性质

一个n阶实方阵若其各阶主子式皆非正,则称为部分非正阵,简写作p.n.p.矩阵.特别地,各阶主子式皆负的p.n.p.矩阵称为部分负矩阵,简写为p.n.矩阵。文[1]、[5]讨论了p.n.p.矩阵的谱性质。本文在[5]的基础上讨论了p.n.p.矩阵的若干性质,并给出p.n.p.矩阵特征值的某些估计式。 引理1 设A=(A_(ij)_n×n为一p.n.p.矩阵,则A的特征值之实部不全为负(n≥2)。 证 设λ_1,λ_2,…,λ_n为A的全部特征值。假定A的每一特征值之实部皆为负。分两种情     

逆p·n·p·矩阵的表征

一个n阶实方阵A,若其各阶主子式皆非正,则称A为p.n.p.矩阵,记作A∈PNP;特别地,若A∈NP且各阶主子式皆负,则称A为p.n.矩阵,记作A∈PN进一步,若n阶实方阵A非奇异,且A^-1∈PNP,则称A为逆p.n.p.矩阵,记作A∈IPNP;特别地,若A^-1∈PN,则称A为逆p.n.矩阵,记作A∈IPN。     

对称正交反对称矩阵反问题解存在的条件

矩阵反问题和矩阵特征值反问题在科学和工程技术中具有广泛的应用,有关它们的研究已取得了许多进展[1,2].[3]和[4]分别研究了反对称矩阵反问题和双反对称矩阵特征值反问题等.本文研究一类更广泛的对称正交反对称矩阵反问题.用Rn×m（Cn×m）表示n×m实（复）矩阵的全体,ASRn×n表示n阶反对称矩阵的全体,ABSRn×n表示n阶双反对称矩阵的全体,ORn×n表示n阶正交矩阵的全体.A+表示矩阵A的Moore-Penrose广义逆.In表示n阶单位矩阵.ei表示n阶单位矩阵的第i列,Sn=[en,en-1,     

一类特殊矩阵的极值特征值反问题

针对矩阵特征值反问题,如何构造矩阵显得尤为重要,鉴于此,引入一种新的带比例关系矩阵.结果表明,只需利用其顺序主子阵的最小和最大特征值即可反构原矩阵,同时亦总结了矩阵元素与顺序主子阵特征值的关系.     

实对称五对角矩阵的两类广义特征值反问题

讨论了如下两类广义特征值反问题:(i)由给定的三个互异的特征对和给定的实对称正定五对角矩阵构造一个实对称五对角矩阵;(ii)由给定的三个互异特征对和给定的全对称正定五对角矩阵构造一个全对称五对角矩阵.利用线性方程组理论、对称向量和反对称向量的性质,分别得到了两类反问题存在唯一解的充要条件,并给出了解的表达式和数值算法;最后通过数值例子说明了算法的有效性.     

基于反Krylov矩阵正交分解的半可分矩阵

在本文中，我们证明了对一个反Krylov矩阵作QR分解后，利用得到的正交矩阵可以将一个具有互异特征值的对称矩阵转化为一个半可分矩阵的形式，这个结果表明了反Krylov矩阵与半可分矩阵之间的联系．另外，我们还证明了这类对称半可分矩阵在QR达代下矩阵结构保持不变性．     

斜对称双对角矩阵特征值反问题

讨论了关于斜对称双对角矩阵的特征值反问题．即：已知一个n阶斜对称双对角矩阵的特征值和两个n-1阶子矩阵的部分特征值，则可求得该矩阵．最后给出了数值例子．     

带比例关系的实带状矩阵特征值反问题

研究了通过矩阵A的顺序主子矩阵A_((k))=(aij)_(i,j=1)^(n-k+1)的特征值{λ_i(n-k+1)的特征值{λ_i^((k)))}_(i=1)((k)))}_(i=1)^(n-k+1)k=1,2,…,r+1来构造一个带比例关系的实带状矩阵的特征值反问题.对当特征值{λ_i(n-k+1)k=1,2,…,r+1来构造一个带比例关系的实带状矩阵的特征值反问题.对当特征值{λ_i^((k))}_(i=1)((k))}_(i=1)^(n-k+1)中有多重特征值出现时,应当如何来构造这类矩阵进行了讨论,并给出了问题的具体算法及数值例子.     

由谱数据和主子矩阵构造Jacobi矩阵及其推广

研究由主子矩阵和谱数据构造Jacobi矩阵问题, 推广到弹簧质点系统谱约束下的双倍维扩展问题和对称三对角二次束逆特征值问题.给出了问题的数值解法.     

关于Jacobi矩阵逆特征值问题的扰动分析

１预备 若不特别说明,本文沿用［６］中记号． Ｈｏｃｈｓｔａｄｔ于１９６７年提出如下问题［１］： 问题Ⅰ　给定两组实数｛λ｝ｎｊ＝１＝１和｛μ｝ｎ＝１ｉ＝１,满足构造一个ｎ阶实对称三对角矩阵Ｊｎ,使得λ１,…λｎ为人的特征值,而Ｊｎ－１阶顺序主子阵的特征值为μ１,…,μｎ－１． 问题Ⅱ　给定一组实数｛λｊ｝ｎｊ＝１,满足构造一个ｎ阶全对称三对角矩阵Ｊｎ（ｓ）,使得Ｊｎ（ｓ）的特征值为λ１,λ２,…λｎ． ｄｅ Ｂｏｏｒ和Ｇｏｌｕｂ［４］提出如下问题： 问题Ⅲ　给定两组实数满足构造ｎ阶实对称三对角矩阵Ｊ…     

Submatrices of non-tree-realizable distance matrices

We deal with distance matrices of real (this means, not necessarily integer) numbers. It is known that a distance matrix D of order n is tree-realizable if and only if all its principal submatrices of order 4 are tree-realizable. We discuss bounds for the number, denoted Q_i(D), of non-tree-realizable principal submatrices of order i ? 4 of a non-tree-realizable distance matrix D of order n?i, and we construct some distance matrices which meet extremal conditions on Q_i(D). Our starting point is a proof that a non-tree-realizable distance matrix of order 5 has at least two non-tree-realizable principal submatrices of order 4. Optimal realizations (by graphs with circuits) of distance matrices which are not tree-realizable are not yet as well known as optimal realizations which are trees. Using as starting point the optimal realization of the (arbitrary) distance matrix of order 4, we investigate optimal realizations of non-tree-realizable distance matrices with the minimum number of non-tree-realizable principal submatrices of order 4.     

Extremal inverse eigenvalue problem for symmetric doubly arrow matrices

In this paper, we consider the following inverse eigenvalue problem: to construct a real symmetric doubly arrow matrix A from the minimal and maximal eigenvalues of all its leading principal submatrices. The necessary and sufficient condition for the solvability of the problem is derived. We also give a necessary and sufficient condition in order that the constructed matrices can be nonnegative. Our results are constructive and they generate algorithmic procedures to construct such matrices.     

Completely normal matrices

Normal matrices in which all submatrices are normal are said to be completely normal. We characterize this class of matrices, determine the possible inertias of a particular completely normal matrix, and show that real matrices in this class are closed under (general) Schur complementation. We provide explicit formulas for the Moore–Penrose inverse of a completely normal matrix of size at least four. A result on irreducible principally normal matrices is derived as well.     

A necessary and sufficient condition for the stabilisation of a matrix and its principal submatrices

We present a necessary and sufficient condition for the existence of a real diagonal matrix such that its product with a given real square matrix is Hurwitz together with all its principal submatrices. We also give an explicit formula for a matrix which, in particular, supplies all the involved matrices with negative distinct eigenvalues.     

Interlacing properties of the eigenvalues of some matrix classes

We establish the eigenvalue interlacing property (i.e. the smallest real eigenvalue of a matrix is less than the smallest real eigenvalue of any of its principal submatrices) for the class of matrices introduced by Kotelyansky (all principal and almost principal minors of these matrices are positive). We show that certain generalizations of Kotelyansky and totally positive matrices possess this property. We also prove some interlacing inequalities for the other eigenvalues of Kotelyansky matrices.     

Characterization of supercommuting matrices

For a given n-by-n matrix A, we consider the set of matrices which commute with A and all of whose principal submatrices commute with the corresponding principal submatrices of A. The properties of this set are examined, with particular attention to its dimension.     

Recognizing Balanceable Matrices

A 0/±1 matrix is balanced if it does not contain a square submatrix with exactly two nonzero entries per row and per column in which the sum of all entries is 2 modulo 4. A 0/1 matrix is balanceable if its nonzero entries can be signed ±1 so that the resulting matrix is balanced. A signing algorithm due to Camion shows that the problems of recognizing balanced 0/±1 matrices and balanceable 0/1 matrices are equivalent. Conforti, Cornuéjols, Kapoor and Vušković gave an algorithm to test if a 0/±1 matrix is balanced. Truemper has characterized balanceable 0/1 matrices in terms of forbidden submatrices. In this paper we give an algorithm that explicitly finds one of these forbidden submatrices or shows that none exists. Received: October 2004     

Bapat-Kwong矩阵不等式的推广

本文研究了两个经典的Hermitian正定矩阵的Hadamard乘积的Bapat-Kwong矩阵不等式的推广,利用局部完全Hermitian矩阵的性质,根据可逆矩阵的主子矩阵与其Schur补的关系,得到了两个局部完全Hermitian矩阵的Hadamard乘积的矩阵不等式.所得到的结果不仅在放弃了正定性的前提下得到了经典的Bapat-Kwong矩阵不等式,而且还给出了这个矩阵不等式等式成立的充分必要条件.     

Eight expressions for generalized inverses of a bordered matrix

A bordered matrix is a two-by-two partitioned matrix with its lower-right corner equal to a null matrix. In this article, we present eight partitioned matrices consisting of the Moore–Penrose inverses of submatrices in a bordered matrix, and give necessary and sufficient conditions for the eight partitioned matrices to be generalized inverses of the bordered matrix.