Error estimation on the Padé approximation of transfer functions via the Lanczos process (Q1279546)

A new derivation of the PVL algorithms (Padé approximation via the Lanczos process) is presented from the matrix point of view in order to simplify the mathematical theory and derivation of the algorithm. There is also given an explicit formulation of the approximation error. This error expression enables implement of PVL algorithm that adaptively determines the number of Lanczos steps required to satisfy a prescribed error tolerance.