Pattern recognition in the Athens stock market: further evidence using a non-parametric approach

A nonparametric test like stochastic dominance analysis, which relies on cumulative densities of observed returns, requires no assumptions regarding the nature of underlying return distributions and imposes few restrictions on investor utility functions. Therefore, this study uses stochastic dominance comparisons to audit previous parametric tests of the day of the week anomaly in Athens Stock Exchange between 1985 and 2004. The results of stochastic dominance analysis, by taking into consideration the thin trading that is common in such a market, show that the day effect is a fact in ASE.