ANALYZING THE FRACTAL STRUCTURE OF STOCK RETURNS: EVIDENCE FROM ISTANBUL STOCK EXCHANGE

Authors : Mehmet Horasanli

Pages : 243-249

Doi:10.14783/maruoneri.679718

View : 18 | Download : 8

Publication Date : 2008-06-10

Article Type : Research

Abstract :Efficient Market Hypothesis States that alt new information is reflected in the market price fully and immediately. Security returns are essentially unpredictable since they follow a random walk. Therefore the impact of the new information is essentially unpredictable; it is as likely to be negative as positive. Financial asset returns are often modeled with a series of small, normally distributed changes. Brownian motion asserts the independence of the changes but there are patterns or trends in Capital market returns and they persist over time. Therefore security returns are not fully random. This paper applies Hurst’s R/S (Rescaled Range) analysis to XU030 and XU100 index within different time horizons. The analysis proceeded from two basic principles: dependence of each period in time series data and fractional Brownian motion of time series. The persistence behaviour of İstanbul Stock Exchange is investigated. The results show that each series taken into consideration exhibits a biased process characteristic of fractal Brownian motion.
Keywords : Brownian Motion, Fractal Brownian Motion, Hurst Exponent, Rescaled Range Analysis