A COMPARATIVE STUDY OF FastICA AND GRADIENT ALGORITHMS FOR STOCK MARKET ANALYSIS

• Authors: Nermend Kesra , Yasen Rajihy
• Languages of publication: EN

Abstracts

EN

In this paper we proved that a fast fixed point algorithm known as FastICA algorithm depending on maximization the nongaussianity by using the ne-gentropy approach is one of the best algorithm for solving ICA model. We compare this algorithm with Gradient algorithm. The Abu Dhabi Islamic Bank (ADIB) used as illustrative example to evaluate the performance of these two algorithms. Experimental results show that the FastICA algorithm is more robust and faster than Gradient algorithm in stock market analysis.

Keywords

EN

independent component analysis; nangaussianity; negentropy; stock market analysis

• Publisher: Katedra Ekonometrii i Statystyki, Szkoła Główna Gospodarstwa Wiejskiego w Warszawie
• Journal: Metody Ilościowe w Badaniach Ekonomicznych
• Year: 2014
• Volume: 15
• Issue: 1
• Pages: 142-152

Dates

published

2014

Contributors

author

Nermend Kesra

• Institute of IT in Management, University of Szczecin, Poland

author

Yasen Rajihy

• Department of Computer Science, University of Babylon, Iraq

References

• Belouchrani A., Abed Meraim K., Cardoso J.F. and Moulines E. (1996) A blind source se-paration technique using second order statistics, IEEE Trans. Signal Processing, vol. 45, 434–444.
• Cardoso J.F., Souloumiac A. (1993) Blind beam forming for non-Gaussian signals, IEEE Proceedings-F, 140(6), 362-370.
• Choi S., Cichocki A., Deville Y. (2001) Differential decorrelation for nonstationary source separation, [in:] Proceedings of 3rd International Conference on Independent Compo-nent Analysis and Blind Signal Separation, (San Diego, California, USA), 319-322.
• Comon P. (1994) Independent component analysis — a new concept?, Signal Processing, 36, 287–314.
• Hyvarinen A. (1997) A family of fixed-point algorithms for independent component Ana-lysis, [in:] Proc. ICASSP, Munich, Germany, April 20-24, 3917–3920.
• Hyvärinen A. (1999) Fast and Robust Fixed-Point Algorithms for Independent Component Analysis, IEEE Trans. on Neural Networks, 10(3), 626-634.
• Hyvarinen A. (1999) Survey of Independent Component Analysis, Neural Computing Surveys, 2, 150, 1-36, 94-128.
• Hyvarinen A., Oja E. (1997) A fast fixed-point algorithm for independent component ana-lysis, Neural Computation, 9(7), 1483–1492.
• Hyvärinen A., Oja E. (2000) Independent Component Analysis: Algorithms and Applica-tions, Neural Networks, 13(4-5), 411-430.
• Hyvärinen A., Karhunen J. and Oja E. (2001) Independent Component Analysis, John Wi-ley & Sons, New York.
• Jones M., Sibson R. (1987) What is projection pursuit? Journal of the Royal Statistical So-ciety, Ser., A150, 1-36.
• Jutten C. (2000) Source separation: from dust till dawn, [in:] Proc. 2nd Int. Workshop on Independent Component Analysis and Blind Source Separation (ICA’ 2000), 15-26.
• Jutten C., Herault J. (1991) Blind separation of sources, part I: An adaptive algorithm ba-sed on neuromimetic architecture, Signal Processing, 24, 1–10.
• Pham D.T., Cardoso J. F. (2000) Blind separation of instantaneous mixtures of nonstatio-nary sources, in Proc. ICA, (Helsinki, Finland), 187–192.
• Pham D.T., Garat P. (1997) Blind separation of mixture of independent sources through a quasimaximum likelihood approach, IEEE Trans. Signal Processing, vol. 45, 1712–1725.