Rorecasting Stock Market Indices with Behavioral Trend Equations

• Autorzy: Piotr Jelonek
• Języki publikacji: EN

Abstrakty

EN

The aim of this article is to present Sornette-Johansen model, its main components and feasible financial applications. In this model the main force that drives stock market indices is the aggregate behaviour of interacting agents who imitate each other's decisions and exchange information. The long-term trend equation used here for the price modelling describes a continuous phase transition between two regimes. One of them represents ordinary trading days, the other precedes stock market crash. The Sornette-Johansen model is based on the notion of the complex system, which is a network of particles (or agents) that repetitively interact. According to this approach, the main concepts which are necessary to forecast changes of long- and middle-term stock market trend are: rational herding, positive feedback, self-organization and critical points. This paper is based on the precursory works of Didier Sornette and Anders Johansen. Both authors have significantly contributed to the development of the theory of complex systems and its applications in physics, economics and social sciences. The article outlines the key steps in derivation of time-dependent trend equations as well as the proposition of estimation method that can be applied in case of noisy data. The first section summarises the main concepts that can be used in order to build a model of a market with endogenous flow of information. The second section describes how the model is derived. In the third section the estimation algorithm and the results obtained for the Polish stock market are presented. (fragment of text)

Słowa kluczowe

EN

Finance; Information science; Statistical methods; Financial mathematics; Stock market; Stock market indices

PL

Finanse; Informatyka; Metody statystyczne; Matematyka finansowa; Giełda papierów wartościowych; Wskaźniki giełdowe

• Czasopismo: Scientific Publications / University of Economics in Katowice
• Rocznik: 2006
• Tom: Mathematical, econometrical and computational methods in finance and insurance
• Strony: 39--50

Twórcy

autor

Piotr Jelonek

• Warsaw School of Economics, Poland

Bibliografia

• Binney J.J., Dowrick N.J., Fisher A.J., Newmann M.E.J.: The Theory of Critical Phenomena - An Introduction to Renormalization Group. Clarendon Press, Oxford 1992.
• Gnaciński P., Makowiec D.: Another Type of Log-periodic Oscillations on Polish Stock Market. Article presented during APFA4 conference, 13-15 November 2003.
• Jelonek P.: Modelowanie endogenicznych zmian trendu metodami fizyki statystycznej (unpublished). Masters thesis at Warsaw School of Economics, Warsaw 2004.
• Johansen A., Ledoit O., Sornette D.: Crashes as Critical Points. "European Journal of Theoretical and Applied Finance", 3(2).
• Johansen A., Sornette D.: Finite-Time Singularity in the Dynamics of the World Population, Economy and Economic Indices. "Physica A" 294 (3-4), Elsevier Science 2001.
• Johansen A., Sornette D., Ledoit O.: Predicting Financial Crashes Using Discrete Scale Invariance. "Journal of Risk" 1999, 1.
• Parsapulous K.E., Vrahatis M.N.: Recent Approach to Global Optimization Problems through Particle Swarm Optimization. "Natural Computing" 1, Kluwer Academic Publishers 2002.
• Sornette D.: Critical Phenomena in Natural Sciences - Chaos Fractals, Self-Organization and Disorder: Concepts and Tools. Springer Verlag, New York 2004.
• Sornette D.: Why Stock Markets Crash? Critical Events in Complex Financial Systems. Princeton University Press, Princeton 2003.
• Sornette D., Johansen A.: Large Financial Crashes. "Physics A", 255 (1-2), Elsevier 1997.
• Sornette D., Johansen A.: Sigificcmce of Log-Periodic Precursors of Financial Crashes. "Quantitative Finance" 1(4), Institute of Physics Publishing 2001.
• Sornette D., Zhou W.-X.: Predictability of Large Future Changes in Complex Systems, http://www.ess.ucls.edu/faculty/somette/pub.complex.asp