×

On graded QR decompositions of products of matrices. (English) Zbl 0855.65036

The author introduces a new method to compute the singular values and vectors of a product of \(n \times n\) matrices. The method is based on the successive factorization \(A_1 A_2 \dots A_m = QRP^T\), where \(Q\) is orthogonal, \(R\) is graded upper triangular, and \(P\) is a permutation matrix.

MSC:

65F20 Numerical solutions to overdetermined systems, pseudoinverses
65F15 Numerical computation of eigenvalues and eigenvectors of matrices
PDF BibTeX XML Cite
Full Text: EuDML EMIS