Abstract: Identification and prediction problem of nonlinear time-series generated by discrete dynamic system is considered via Kernel Method approach. A unified approach to recurrent kernel identification algorithms design is proposed. In such a way a recurrent modification of initial Kernel Method with growing windows is considered. In order to prevent the model complexity increasing under on-line identification, the reduced order model kernel method is proposed and proper recurrent identification algorithms are designed along with conventional regularization technique. Such an approach leads to a new type of Recursive Least-Square? Kernel Method identification algorithms. Finally, the recurrent version of Sliding Window Kernel Method is also developed along with suitable identification algorithms. The proposed algorithm has tracking properties and may be successfully used for on-line identification of nonlinear non-stationary time-series.

Keywords: identification, kernel methods, machine learning, nonlinear model, prediction, recurrent least-squares, support vector machine, time-series

ACM Classification Keywords: G. Mathematics of Computing: G.1 Numerical Analysis: Least squares approximation, nonlinear approximation, G.3 Probability and Statistics: Time series analysis

Link:

KERNEL-BASED METHODS FOR NON-STATIONARY TIME-SERIES IDENTIFICATION AND PREDICTION

Leonid Lyubchyk, Vladyslav Kolbasin

http://foibg.com/ibs_isc/ibs-13/ibs-13-p05.pdf