Analyzing the dual long memory in returns and volatility in Euro Zone crisis countries

This study investigates the dual long memory property in the returns and volatility of the five countries? - Italy, Greece, Ireland, Portugal, Spain - stock exchange markets which went through the Eurozone crises. After the analysis of ARFIMA-FIGARCH model, the variance breaks which are obtained from the modified ICSS (Iterated Cumulative Sums of Squares) method have also been incorporated into the models. This way, the impact of structural breaks on the volatility persistence has been captured.According to the estimation results of the ARFIMA-FIGARCH models with structural breaks and without structural breaks, in all of the markets, effect of shocks on volatility persist for a long time. So, the long memory in volatility decays efficient market hypothesis by deteriorating the price mechanism. By the way, the break analysis points out the fact that, indexes are enourmously affected from the regional or global news. Because of this, investors should more carefully follow the reginal or global news.Existence of dual long memory property in the investigated stock markets contradicts the weak form market efficiency, because existence of dual long memory suggests that future prices can be predicted via analyzing past price movements of the equity. As a result it is possible to make speculative gains and excess profits. In an inefficient market, techniques using past prices to forecast futures prices, such as technical and fundamental analysis may be useful to make excess gains.Keywords: Dual Long Memory, Volatility, ARFIMA-FIGARCH, Efficient Market Hypothesis, ICSS, Kappa-1, Kappa-2