Some bounds for extreme singular values of a complex matrix (Q648187)

Let \(A\) be an \(n\times n\) complex matrix. Suppose that the rows and columns of \(A\) are arranged in such a way that \(r_1(A)\geq r_2(A)\geq \cdots \geq r_n(A),\) \(c_1(A)\geq c_2(A)\geq \cdots \geq c_n(A)\) where \(r_i(A)\) is the Euclidean norm of \(i\)-th row of \(A\) and \(c_j(A)\) is the Euclidean norm of the \(j\)-th column of \(A.\) Let the singular values \(\sigma _i(A)\) of \(A,\) are arranged such that \(\sigma_1(A)\geq \sigma_2(A)\geq \cdots \geq \sigma_n(A)\) and \(\alpha _i\) are positive real numbers such that \(\alpha _1\leq \alpha _2\leq \cdots \leq \alpha _n.\) The authors prove that \[ \sigma_n(A)\leq \frac{\min \{r_1(A),c_1(A)\} \root{n}\of{|\det A|\prod_{i=1}^n\frac{\alpha_i}{\max \{r_i(A),c_i(A)\}}}} {\alpha_1} \] and \[ \frac{\max \{r_n(A),c_n(A)\} \root{n}\of{|\det A|\prod_{i=1}^n\frac{\alpha_i}{\min \{r_i(A),c_i(A)\}}}} {\alpha_n}\leq \sigma_1(A). \] As application, the authors obtain bounds for the extreme singular values of Hilbert and Cauchy-Hankel matrices.