A Bayesian nonlinear support vector machine error correction model

Carine Brasseur, Marcelo Espinoza, Johan A. K. Suykens, Tony van Gestel (), Bart Baesens and Bart De Moor
Additional contact information
Carine Brasseur: Fortis Bank Brussels, Belgium, Postal: Fortis Bank Brussels, Belgium
Marcelo Espinoza: Katholieke Universiteit Leuven, Belgium, Postal: Katholieke Universiteit Leuven, Belgium
Johan A. K. Suykens: Katholieke Universiteit Leuven, Belgium, Postal: Katholieke Universiteit Leuven, Belgium
Bart Baesens: School of Management, University of Southampton, UK, Postal: School of Management, University of Southampton, UK
Bart De Moor: Katholieke Universiteit Leuven, Belgium, Postal: Katholieke Universiteit Leuven, Belgium

Journal of Forecasting, 2006, vol. 25, issue 2, pages 77-100

Abstract: The use of linear error correction models based on stationarity and cointegration analysis, typically estimated with least squares regression, is a common technique for financial time series prediction. In this paper, the same formulation is extended to a nonlinear error correction model using the idea of a kernel-based implicit nonlinear mapping to a high-dimensional feature space in which linear model formulations are specified. Practical expressions for the nonlinear regression are obtained in terms of the positive definite kernel function by solving a linear system. The nonlinear least squares support vector machine model is designed within the Bayesian evidence framework that allows us to find appropriate trade-offs between model complexity and in-sample model accuracy. From straightforward primal-dual reasoning, the Bayesian framework allows us to derive error bars on the prediction in a similar way as for linear models and to perform hyperparameter and input selection. Starting from the results of the linear modelling analysis, the Bayesian kernel-based prediction is successfully applied to out-of-sample prediction of an aggregated equity price index for the European chemical sector. Copyright © 2006 John Wiley & Sons, Ltd.

Downloads: (external link)
http://hdl.handle.net/10.1002/for.975 Link to full text; subscription required (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Access Statistics for this article

Journal of Forecasting is edited by Derek W. Bunn

More articles in Journal of Forecasting from John Wiley & Sons, Ltd.
Series data maintained by Christopher F. Baum ().