Advertisement

Adaptive Trading for Anti-correlated Pairs of Stocks

  • Chih-Hao LinEmail author
  • Sai-Ping Li
  • K. Y. Szeto
Chapter
  • 1.5k Downloads
Part of the Lecture Notes in Economics and Mathematical Systems book series (LNE, volume 669)

Abstract

The effect of anti-correlation between stocks in real stock market can be exploited for profit if one can also properly set the criterion for trading that takes into account the volatility of the stock pair. This complex problem of resource allocation for portfolio management of stocks is here simplified to a problem of adaptive trading with an investment criterion that evolves along with the time series of the stock data. The trend of the stock is modeled with standard stochastic dynamics, from which the volatility of the stock provides a criterion for investment on a two stock portfolio that consists of the anti-correlated pair using mean variance analysis that optimizes the return. The action of buy and sell of the two-stock portfolio will be based on the fractional return of the pair: when the fractional return of the pair is greater than an upper threshold of 1.01, the action “buy” is taken; and when this fractional return is less than a lower threshold of 0.99, the action “sell” is taken. Since both the volatility criterion for investment and the fractional return of the two-stock portfolio are time dependent, the entire trading scheme is adaptive. Comparison of this evolving strategy of investment with time-average performance of the respective stocks indicates a consistent superiority.

Keywords

Stock Market Fixed Threshold Portfolio Management Adaptive Threshold Financial Time Series 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Markowitz H (1952) Portfolio selection. J Financ 7:77Google Scholar
  2. 2.
    Horasanl M, Fidan N (2007) Portfolio selection by using time varying covariance matrices. J Econ Soc Res 9(2):1Google Scholar
  3. 3.
    Hakansson NH (1971) Multi-period mean-variance analysis: toward a general theory of portfolio choice. J Financ 26:857Google Scholar
  4. 4.
    Maccheroni F (2009) Portfolio selection with monotone mean-variance preferences. Math Financ 19:487CrossRefGoogle Scholar
  5. 5.
    Campbell J, Viceira L (2002) Strategic asset allocation-portfolio choice for long-term investors. Clarendon lectures in economics. Oxford University Press, OxfordCrossRefGoogle Scholar
  6. 6.
    Schweizer M (1995) Variance optimal hedging in discrete time. Math Oper Res 20:1CrossRefGoogle Scholar
  7. 7.
    Blanchet-Scalliet C, El Karoui N, Jeanblanc M, Martellini L (2008) Optimal investment decisions when time-horizon is uncertain. J Math Econ 44:1100CrossRefGoogle Scholar
  8. 8.
    Fukunaga K (1990) Introduction to statistical pattern recognition, 2nd edn. Academic, BostonGoogle Scholar
  9. 9.
    Zemke S (1999) Nonlinear index prediction. Physica A269:177–183Google Scholar
  10. 10.
    Szeto KY, Cheung KH (1998) Multiple time series prediction using genetic algorithms optimizer. In: Proceedings of the international symposium on intelligent data engineering and learning, IDEAL’98, Hong Kong, pp 127–133Google Scholar
  11. 11.
    Szeto KY, Cheung KH (1997) Annealed genetic algorithm for multiple time series prediction. In: Proceedings of the world multiconference on systemic, cybernetics and informatics, Caracas, vol 3, pp 390–396Google Scholar
  12. 12.
    Szeto KY, Chan KO, Cheung KH (1997) Application of genetic algorithms in stock market prediction. In: Weigend AS, Abu-Mostafa Y, Refenes APN (eds) Proceedings of the fourth international conference on neural networks in the capital markets: progress in neural processing, decision technologies for financial engineering, NNCM-96, California. World Scientific, pp 95–103Google Scholar
  13. 13.
    Froehlinghaus T, Szeto KY (1996) Time series prediction with hierarchical raidal basis function. In: Proceedings of the international conference on neural information processing, ICONIP’96, Hong Kong, vol 2. Springer, pp 799–802Google Scholar
  14. 14.
    Szeto KY, Fong LY (2000) How adaptive agents in stock market perform in the presence of random news: a genetic algorithm approach. In: Leung KS et al (eds) IDEAL 2000, Hong Kong. Lecture notes in computer science/Lecture notes in artificial intelligence, vol 1983. Springer, Heidelberg, pp 505–510Google Scholar
  15. 15.
    Fong ALY, Szeto KY (2001) Rule extraction in short memory time series using genetic algorithms. Eur Phys J B 20:569–572CrossRefGoogle Scholar
  16. 16.
    Chen C, Tang R, Szeto KY (2008) Optimized trading agents in a two-stock portfolio using mean-variance analysis. In: Proceedings of the 2008 Winter WEHIA & CIEF, Taipei, pp 61–67Google Scholar
  17. 17.
    Chen W, Szeto KY (2012) Mixed time scale strategy in portfolio management. Int Rev Financ Anal 23:35–40CrossRefGoogle Scholar
  18. 18.
    Bacry E, Delour J, Muzy JF (2001) Multifractal random walk. Phys Rev E 64:026103CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.Institute of PhysicsAcademia SinicaTaipeiChina
  2. 2.Department of PhysicsThe Hong Kong University of Science and TechnologyHong Kong SARChina

Personalised recommendations