Empirical distribution of stock returns of Southeast European emerging markets

The assumption that equity returns follow the normal distribution, most commonly made in financial economics theory and applications, is strongly rejected by empirical evidence presented in this paper. As it was found in many other studies, we confirm that stock returns follow a leptokurtic distribution and skewness, which in most of the Southeast European (SEE) markets is negative. This paper investigates further whether there is any distribution that may be considered an optimal fit for stock returns in the SEE region. Using daily, weekly and monthly data samples for a period of five years from ten Southeast European emerging countries, we applied the Anderson-Darling test of Goodness-of-fit. We strongly rejected the aforementioned assumption of normality for all considered data samples and found that the daily stock returns are best fitted by the Johnson SU distribution whereas for the weekly and monthly stock returns there was not one predominant, but many distributions that can be considered a best fit.