Nevanlinna formula for the truncated matrix trigonometric moment problem (Q352139): Difference between revisions

The author continues his study of the truncated matrix trigonometric moment problem \[ \int_0^{2\pi} e^{int}\,dM(t)=S_n, \qquad n=0,1,\dotsc,d, \] where \(dM\) is an unknown matrix measure, and \(\{S_n\}_{n=0}^d\) are given complex matrices (moments) of order \(N\). The main goal of the paper is the construction of the Nevanlinna formula which provides the complete set of solutions in the indeterminate case (when there are infinitely many such solutions). The underlying operator \(A\) in this case is an isometry with both defect indices equal \(\delta\geq1\). The Nevanlinna formula describes the set of the Cauchy transforms of the measures \(dM\) as a certain linear fractional map \[ \int_0^{2\pi} \frac{dM^T(t)}{1-ze^{it}}=\frac{A_z}{h_z}-z\frac{B_z}{h_z}F_z\left(I_\delta+\frac{C_z}{h_z}F_z\right)^{-1}D_z, \qquad |z|<1, \] where \(A_z\), \(B_z\), \(C_z\), and \(D_z\) are explicitly given matrix polynomials of orders \(N\times N\), \(N\times\delta\), \(\delta\times\delta\), and \(\delta\times N\), respectively, and \(h_z\) is a scalar polynomial. A free parameter \(F_z\) is a \(\delta\times\delta\) matrix valued analytic and contractive function in the unit disk. The correspondence between all such functions and all solutions of the moment problem is bijective. The author shows that the above formula can be adapted to solve the determinate moment problem. A few illuminating examples are considered.