DocumentCode :
3262776
Title :
Cointegration by MCA and modular MCA
Author :
Xu, Lei ; Leung, Wai Man
Author_Institution :
Dept. of Comput. Sci. & Eng., Chinese Univ. of Hong Kong, Shatin, Hong Kong
fYear :
1998
fDate :
29-31 Mar 1998
Firstpage :
157
Lastpage :
160
Abstract :
A new approach for investigating the cointegration relationship between time series is proposed based on the so called minor component analysis (MCA) and its extension, modular MCA. With no need to place one variable on the left-hand side and use the others as regressors on the estimation of the long-run equilibrium regression, and also no need to build up a vector autoregression model for rank based analysis, this approach conducts cointegration analysis directly based on the linear combination of the instantaneous values of nonstationary variables via an eigen-analysis corresponding to the smallest eigenvalue, which can be implemented either in a batch way or by a Hebbian learning adaptive algorithm. A method for estimating the order of the cointegration has been also developed. Experiments are made in comparison with the least square regression based cointegration approach and have demonstrated the advantages of the approach
Keywords :
Hebbian learning; eigenvalues and eigenfunctions; financial data processing; integration; least squares approximations; statistical analysis; time series; Hebbian learning adaptive algorithm; MCA; batch method; cointegration; eigenanalysis; eigenvalue; finance; instantaneous nonstationary variable values; least square regression based cointegration approach; long-run equilibrium regression; minor component analysis; modular MCA; modular minor component analysis; time series; Adaptive algorithm; Computer science; Eigenvalues and eigenfunctions; Hebbian theory; Least squares methods; Reactive power; Testing; Time series analysis; Vectors; White noise;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computational Intelligence for Financial Engineering (CIFEr), 1998. Proceedings of the IEEE/IAFE/INFORMS 1998 Conference on
Conference_Location :
New York, NY
Print_ISBN :
0-7803-4930-X
Type :
conf
DOI :
10.1109/CIFER.1998.690071
Filename :
690071
Link To Document :
بازگشت