The Chaotic Nature of Financial Markets
Volume 15, Number 2 Article by Ajit Haridas June, 2003
Award Winning Student Essay Order in Disorder: The Chaotic Nature of Financial Markets :
The simplicity of the Capital Market Theory makes it ideal for an elementary appreciation of finance. Analysts have traditionally used the Efficient Market Hypothesis (EMH) and its variations to explain market inconsistencies, though the EMH itself was based on certain questionable assumptions such as a normal distribution of returns and the concept of a rational investor. Financial markets, however, involve nonlinear structures, growth and development, and such linear tools are inadequate for modelling them. Non-linear relationships such as the one between the real interest rate and the money supply, exhibit interdependence between variables that can lead to very complex relationships. In order to understand the intricate relationships between multiple interdependent variables, we must resort to the chaos theory.
This award winning student essay by Ajit Haridas explores the possibility of adopting a new paradigm for understanding the behaviour of financial markets. The Fractal Market Hypothesis (FMH), he finds, utilises chaos theory along with fractal geometry to provide a more realistic model. In a dynamic system, the time-related or temporal element plays a crucial role. Fractals and chaos, while modelling different phenomena, are both nonlinear dynamic feedback systems with a memory element where the current outputs act as inputs to the next state. In regard to financial markets, this memory element allows markets to take into account past events. Comparing the EMH and the FMH on the four parameters of efficiency, investment horizons, market memory and the validity of the normal distribution, Haridas finds the latter a much more appropriate model for analysing the phenomena of the financial market. Finally, he discusses a few implications of this new model for econometrics and financial forecasting. The most difficult aspect of prediction is identifying the variables that affect the system and finding the relationship between those variables. Neural networks present us with a possible solution. Once enough inputs are gathered to characterise the system, the neural net should succeed in developing a relationship. An investor who succeeds in figuring out the relationship between interdependent market variables will be at a significant advantage over other market participants. The key lies in arriving at that elusive equation.
Reprint No 03202