Summary

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

2012

Session Number:C1L-A

Session:

Number:523

Stock Portfolio Management Based on Nonlinear Prediction Model

Satoshi Inose,  Tomoya Suzuki,  Kazuo Yamanaka,  

pp.523-526

Publication Date:

Online ISSN:2188-5079

DOI:10.15248/proc.1.523

PDF download (377.6KB)

Summary:
A re-formulation of the Markowitz mean-variance portfolio model is attempted in order to make it suitable to more complicated situations of practical financial markets. An AR model with data-dependent coefficients are introduced for predicting the future return in place of the simpler arithmetic mean of past data. This means that a class of nonlinear predictor is used. The prediction error variance also replaces the variance of past data as the evaluation index for the risk. A computer simulation based on practical data on stock prices suggests that investment by using the new portfolio model results in higher profit with lower risk.

References:

[1] H. M. Markowitz: “Portfolio Selection,” Journal of Finance, Vol.7, No.1, pp.77-91, 1952.

[2] E. F. Fama: “Efficient capital markets: A review of theory and empirical work,” Journal of Finance, pp.383-417, 1970.

[3] J. Y. Campbell, A. W. Lo, and A. C. Mackinlay: The Econometrics of Financial Markets, Princeton University Press, 1996.

[4] C. Gourieroux and J. Jasiak: Financial Econometrics: Problems, Models, and Methods., Princeton University Press, 2001.

[5] F. Black and R. Litterman: “Global Portfolio Optimization,” Financial Analysts Journal, pp.28-43, 1992.

[6] W. F. Sharpe: “Capital asset prices: A theory of market equilibrium under conditions of risk,” Journal of Finance, Vol.19, No.3, pp.425-442, 1964.

[7] The historical data of Overnight unsecured call money was obtained from The Tokyo Tanshi Co., Ltd., http://www.tokyotanshi.co.jp/past/index2.shtml, January 15, 2012.

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

[9] The historical stock data was bought from Pan Rolling Inc, http://www.panrolling.com/pansoft/data/, July 30, 2011.