Robust time series estimation
Demetrios Kazakos, Kami S. Makki · International Conference on System Science and Simulation in Engineering · 2006
In this paper we present some new results on the problem of robust estimation for stationary multiple time series processes. For these processes, we consider the prediction, smoothing and causal filtering problem in cases for which the minimum achievable mean square error is expressed in a closed form in terms of the spectral density matrix of the signal. We consider three convex classes of spectral uncertainties, and develop robust solutions for these cases.