The inverse eigenvalue problem for symmetric anti-bidiagonal matrices (Q2568376): Difference between revisions

The goal of this paper is the inverse eigenvalue problem for symmetric anti-bidiagonal matrices, that is to characterize the spectra of matrices of the form \[ A=\left[ \begin{matrix} 0 & 0 &\cdots & 0 & a_n \\ 0 & 0 &\cdots &a_{n-2}&a_{n-1}\\ \vdots&\vdots& &\vdots&\vdots \\ 0&a_{n-2}& \cdots &0 & 0 \\ a_n & a_{n-1} & \cdots & 0 & 0 \end{matrix} \right], \] where all \(a_j\) are positive. The author shows that a real \(n\)-tuple \(\Lambda =(\lambda_1,\ldots,\lambda_n)\) can be realized as the spectrum of a symmetric bidiagonal matrix with all \(a_j\) positive if and only if \[ \lambda_1 >-\lambda_2>\lambda_3> \cdots >(-1)^{n-1}\lambda_n. \] The realizing matrix is shown to be unique. The author also proves that this problem is equivalent to the inverse eigenvalue problem for a certain subclass of Jacobi matrices.