Products of idempotent matrices

For some years it has been known that every singular square matrix over an arbitrary field F is a product of idempotent matrices over F. This paper quantifies that result to some extent. Main result: for every field F and every pair (n,k) of positive integers, an n×n matrix S over F is a product of k idempotent matrices over F iff rank(I ? S)?k· nullity S. The proof of the “if” part involves only elementary matrix operations and may thus be regarded as constructive. Corollary: (for every field F and every positive integer n) each singular n×n matrix over F is a product of n idempotent matrices over F, and there is a singular n×n matrix over F which is not a product of n ? 1 idempotent matrices.     

Products of EP matrices

Necessary and sufficient conditions are given for a matrix to be a product of an EP_r matrix by an EP_s matrix. It is shown that a given square matrix is a product of more than two EP matrices of specified ranks (and hence nullities) if and only if its rank is less than or equal to the minimum of the given ranks and its nullity is less than or equal to the sum of the given nullities. It is also shown that given two EP matrices, the rank of their product is independent of the order of the factors.     

Products of nilpotent matrices

We show that any complex singular square matrix T is a product of two nilpotent matrices A and B with rank A = rank B = rank T except when T is a 2×2 nilpotent matrix of rank one.     

Products of irreducible matrices

In this paper we characterize the subsemigroup of B_n (B_n is the multiplicative semigroup of n × n Boolean matrices) generated by all the irreducible matrices, and hence give a necessary and sufficient condition for a Boolean matrix A to be a product of irreducible Boolean matrices. We also give a necessary and sufficient condition for an n × n nonnegative matrix to be a product of nonnegative irreducible matrices.     

On matrices stochastically similar to matrices with equal diagonal elements

A necessary and sufficient condition for a matrix to be stochastically similar to a matrix with equal diagonal elements is obtained Aand B are called Stochastically similar if B=SAS^{- 1} where S is quasi-stochastic i.e., all row sums of .S are I. An inverse elementary divisor problem for quasi-stochastic matrices is also considered.     

On matrices stochastically similar to matrices with equal diagonal elements

A necessary and sufficient condition for a matrix to be stochastically similar to a matrix with equal diagonal elements is obtained Aand B are called Stochastically similar if B=SAS ^{? 1} where S is quasi-stochastic i.e., all row sums of .S are I. An inverse elementary divisor problem for quasi-stochastic matrices is also considered.     

On complex matrices that are unitarily similar to real matrices

There are well-known conditions ensuring that a complex n × n matrix A can be converted by a similarity transformation into a real matrix. Is it possible to realize this conversion via unitary similarity rather than a general one? The following answer to this question is given in this paper: A matrix A ∈ M _n (?) can be made real by a unitary similarity transformation if and only if A and ā are unitarily similar and the matrix P transforming A into ā can be chosen unitary and symmetric at the same time. Effective ways for verifying this criterion are discussed.     

On complex matrices with simple spectrum that are unitarily similar to real matrices

Suppose that one should verify whether a given complex n × n matrix can be converted into a real matrix by a unitary similarity transformation. Sufficient conditions for this property to hold were found in an earlier publication of this author. These conditions are relaxed in the following way: as before, the spectrum is required to be simple, but pairs of complex conjugate eigenvalues $ \lambda ,\bar \lambda $ \lambda ,\bar \lambda are now allowed. However, the eigenvectors corresponding to such eigenvalues must not be orthogonal.     

Products of involutory matrices. I

It is shown that, for every integer ≥1 and every field F, each n×n matrix over F of determinant ±1 is the product of four involutory matrices over F. Products of three ×n involutory matrices over F are characterized for the special cases where n≤4 or F has prime order ≤5. It is also shown for every field F that every matrix over F of determinant ±1 having no more than two nontrivial invariant factors is a product of three involutory matrices over F.     

Products of involutory matrices. I

It is shown that, for every integer ?1 and every field F, each n×n matrix over F of determinant ±1 is the product of four involutory matrices over F. Products of three ×n involutory matrices over F are characterized for the special cases where n?4 or F has prime order ?5. It is also shown for every field F that every matrix over F of determinant ±1 having no more than two nontrivial invariant factors is a product of three involutory matrices over F.     

Products of elementary doubly stochastic matrices

While studying a theorem of Westwerk on higher numerical ranges, we became interested in how the theory of elementary doubly stochastic (e.d.s.) matrices is related to a result of Goldberg and Straus. We show that there exist classes of doubly stochastic (d.s.) matrices of order n≧3 and orthostochastic (o s) matrices of order n≧4 such that the matrices in these classes cannot be represented as a product of e.d.s. matrices. In fact the matrices in these classes do not admit a representation as an infinite limit of a product of e.d.s. matrices.     

Products of complic cosquares and pseudo-involutory matrices

LetF be a field with (nontrivial) involution (i.e.F-conjugation). A nonsingular matrix Aover Fis called a complic F-cosquare provided A=S^*-1for some matrix Sover Fand is called p.i. (pseudo-involutory) provided A=A^-1 It is shown that Ais a complic F-cosquare iff Ais the product of two p.i. matrices over Fand that det (AA)=1 iff Ais the product of two complic F-cosquares (hence iff A is the product of four p.i. matrices over F). It is conjectured that, except for one obvious case (2 x 2 matrices over the field of order 2), every unimodular matrix A over an arbitrary field Fis a product S¹ST:1T with S1 and Tover FThis conjecture is proved for matricesAof order ≤3.     

Products of two involutory matrices over skewfields

If a simple transformation σ is a product of two involutions, then σ is a reflection or a transvection. This property is still true for matrices over skewfields. It will be used to show that a criterion for the decomposability of a matrix into two involutions, which is known for matrices over commutative fields, is no longer true if the entries of the matrix are taken from a skewfield. Another consequence is that the special linear group of a vector space over the field K is not bireflectional if K is not commutative.     

Products of idempotent matrices over integral domains

We say that a ring R has the idempotent matrices property if every square singular matrix over R is a product of idempotent matrices. It is known that every field, and more generally, every Euclidean domain has the idempotent matrices property. In this paper we show that not every integral domain has the idempotent matrices property and that if a projective free ring has the idempotent matrices property then it must be a Bezout domain. We also show that a principal ideal domain has the idempotent matrices property if and only if every fraction a/b with b≠0 has a finite continued fraction expansion. New proofs are also provided for the results that every field and every Euclidean domain have the idempotent matrices property.     

Products of Beta matrices and sticky flows

A discrete model of Brownian sticky flows on the unit circle is described: it is constructed with products of Beta matrices on the discrete torus. Sticky flows are defined by their moments which are consistent systems of transition kernels on the unit circle. Similarly, the moments of the discrete model form a consistent system of transition matrices on the discrete torus. A convergence of Beta matrices to sticky kernels is shown at the level of the moments. As the generators of the n-point processes are defined in terms of Dirichlet forms, the proof is performed at the level of the Dirichlet forms. The evolution of a probability measure by the flow of Beta matrices is described by a measure-valued Markov process. A convergence result of its finite dimensional distributions is deduced.Mathematics Subject Classification (2000):60J27, 60J35, 60G09     

Products of idempotent matrices over Hermite domains

Communicated by F. Pastijn     

Products of two unipotent matrices of index 2

Necessary and sufficient conditions are presented for a square matrix A over a general field F to be the product of two unipotent matrices of index 2. This generalizes a result established by Wang and Wu (1991) [4] for the case where F is the complex field.     

Products of matrices with prescribed spectra and ranks

This paper studies the possibility of writing a given square matrix as the product of two matrices with prescribed spectra and ranks. It extends some previously known results.