✍ Scribed by T. Söderström; P. Stoica
No coin nor oath required. For personal study only.
Aimtraet--The output error identification method is studied in various respects. The stationary points of the associated loss function are investigated. Sufficient conditions for a unique local minimum are given. The loss functions can be minimized using a quasilinearization algorithm. Such an algorithm will give good local convergence. It is, however, shown that giobal convergence does not always occur. The output error method is also compared with some other estimation methods from the accuracy point of view. It is proved that a prediction error method will give better accuracy. An instrumental variable technique may give better or worse accuracy depending on the actual noise correlation.
📜 SIMILAR VOLUMES
When the SPR condition is not satisfied then there exist input signals for which the considered output error algorithm is locally unstable. A condition is given which delimits the sharp stability-instability boundary in the case of slow estimation, whereas local stability properties are guaranteed b