Modeling Chaotic Behaviours In Financial Markets
Lack of deef trading volüme, increasing levels of hot Money and information, flexibilities of hedge funds an volatilitiesin emerging markets interrupt the linear relationship between risk and return. Due to corupted risk perspectives and irrational risk appettites of the market participants, chaotic patterns and non-linear behaviours in financial time series might ocur in emerging markets. Traditional econometric models are not able to capture chaotic natüre of the markets due to their strick assumptions on time series such as requirement of normal distribution fort he series. Recently, artificial neural networks, wavelets, fuzzy logic and genetic algortihms have been uset to model chaotic behaviours. This paper discusses the reasons of emergence of chaotic patterns and algorithms of modeling of those patterns. Neural networks and wavelets are introduced as modeling methods with a simple simulations based on feedforward neural networks. The paper concludes that successfully designed hybrid intelligent models might capture the chaos and non-linearities in the markets.