On Faddeev-Leverrier's method for the computation of the characteristic polynomial of a matrix and of eigenvectors (Q2365725)

A method is presented to compute the eigenvalues and eigenvectors of a matrix \(A\). By this method the characteristic polynomial is obtained as proposed by \textit{D. K. Faddeev} and \textit{V. N. Faddeeva} [Computational methods of linear algebra (1963; Zbl 0094.110), \S 47; 4] which is an improved modification of the method by U. J. J. Leverrier (1840). Necessary and sufficient conditions for \(C(\lambda_ 0)\neq 0\;(C(\lambda):=\sum^{n-1}_{k=0}\lambda^{n-1-k}\bar A_ k)\) are found. It is shown how the derivatives of \(C\) at \(\lambda_ 0\) still may serve to find eigenvectors and generalized eigenvectors of \(A\) if \(C(\lambda_ 0)\) vanishes. A more complete account of the relation between the spans of \(C(\lambda_ 0)\) and of its derivatives \(C^{(m)}(\lambda_ 0)\) at an eigenvalue \(\lambda_ 0\) on the one hand, and the generalized eigenspace of \(A\) for this eigenvalue on the other hand, is given.