Nonlinear Cointegration Analysis Based on Support Vector Machines and Its Application to the Connection between Financial Markets

DOI

10.2991/cse.2013.28How to use a DOI?

Keywords

non-stationary time series; nonlinear cointegration analysis; support vector machine; neural network; the connection between financial markets

Abstract

With the purpose of analyzing non-stationary time series, this paper introduced a novel nonlinear cointegration discriminate analysis by employing support vector machines to overcome existing limitations of two methods, that is, the statistical approach and the neural network used by the nonlinear cointegration theory. The proposed nonlinear cointegration discriminate analysis can effectively address the key step in the nonlinear cointegration test. Then, the application of the innovated method in the investigation of the connection between financial markets was explored. The empirical analysis demonstrates that the support vector machine is significant in investigating the money demand function and its stability and has advantages in dealing with nonlinear cointegration relations among different financial variables and estimating nonlinear function over the two methods above.

Copyright

© 2013, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chuanhe Shen
AU  - Zhongwen Liu
AU  - Ying Li
PY  - 2013/07
DA  - 2013/07
TI  - Nonlinear Cointegration Analysis Based on Support Vector Machines and Its Application to the Connection between Financial Markets
BT  - Proceedings of the 2nd International Conference on Advances in Computer Science and Engineering (CSE 2013)
PB  - Atlantis Press
SP  - 119
EP  - 123
SN  - 1951-6851
UR  - https://doi.org/10.2991/cse.2013.28
DO  - 10.2991/cse.2013.28
ID  - Shen2013/07
ER  -