Kernel based partially linear models and nonlinear identification
- 10 October 2005
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 50 (10), 1602-1606
- https://doi.org/10.1109/tac.2005.856656
Abstract
In this note, we propose partially linear models with least squares support vector machines (LS-SVMs) for nonlinear ARX models. We illustrate how full black-box models can be improved when prior information about model structure is available. A real-life example, based on the Silverbox benchmark data, shows significant improvements in the generalization ability of the structured model with respect to the full black-box model, reflected also by a reduction in the effective number of parameters.Keywords
This publication has 11 references indexed in Scilit:
- Semiparametric Non-Linear Time Series Model SelectionJournal of the Royal Statistical Society Series B: Statistical Methodology, 2004
- Partially linear models and least squares support vector machinesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2004
- Fast approximate identification of nonlinear systemsAutomatica, 2003
- Structural Modelling with Sparse KernelsMachine Learning, 2002
- Least Squares Support Vector MachinesPublished by World Scientific Pub Co Pte Ltd ,2002
- Partially Linear ModelsContributions to Statistics, 2000
- Wavelet Based Modeling of Nonlinear SystemsPublished by Springer Science and Business Media LLC ,1998
- Nonlinear black-box modeling in system identification: a unified overviewAutomatica, 1995
- Spline Models for Observational DataPublished by Society for Industrial & Applied Mathematics (SIAM) ,1990
- Root-N-Consistent Semiparametric RegressionEconometrica, 1988