Summary

Proceedings of the 2012 International Symposium on Nonlinear Theory and its Applications

2012

Session Number:C1L-A

Session:

Number:531

New Bollinger bands for Nonlinear Technical Analysis of Pairs Trading

Taiga Hayashi,  Tomoya Suzuki,  

pp.531-534

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.1.531

PDF download (618.2KB)

The Bollinger bands are well known as one of the technical measures for pairs trading, and are used for detecting excessive price difference between two stocks. These bands mean the probability distribution of a price difference, but actually they have been estimated by simply aggregating historical price differences. In this study, to improve the estimation accuracy of the distribution, we apply the Bagging algorithm based on a nonlinear prediction following local spatial dynamics. Through some investment simulations using real stock prices, we demonstrate that our proposed method is more useful than the conventional Bollinger bands.

[1] J. J. Murphy, “Technical Analysis of the Financial Markets,” Prentice Hall Press, 1999.

[2] J. Bollinger, “Bollinger on Bollinger Bands,” McGraw-Hill, 2001.

[3] K. Nakata and T. Suzuki, “Evaluating the Risk of Nonlinear Prediction with the Bagging Algorithm,” Proc. of NCSP, pp.748-751, 2012.

[4] L. Breiman, “Bagging Predictor,” Machine Learning, Vol.24, pp.123-140, 1996.

[5] J. D. Farmer and J. J. Sidorowich, “Predicting chaotic time series,” Phys. Rev. Lett., Vol.59, pp.845-848, 1987.

[6] F. Takens, “Detecting Strange Attractors in Turbulence,” Lecture Notes in Mathematics, Vol.898, pp.366-381, Springer-Verlag, 1981.

[7] D. Haraki, T. Suzuki, H. Hashiguchi, T. Ikeguchi, “Bootstrap Nonlinear Prediction,” Phys. Rev. E, Vol.75, 056212, 2007.