Analyzing the Dual Long Memory in Stock Market Returns

The purpose of this study is to examine the dual long memory properties for five stock market returns by using joint ARFIMA-FIGARCH model and structural break test in context of weak form efficient market hypothesis. The models are estimated by using daily closing prices for S&P500, FTSE100, DAX, CAC40 and ISE100. In an effort to assess the impact of structural breaks in volatility persistence, the breaks in variance are detected by using the Iterated Cumulative Sums of Squares (ICSS) algorithm, and dummy variables are incorporated to the models. Empirical findings show that the dual long memory exists for all stock markets. Also the volatility has a predictable structure and indicates that all stock markets are weak form inefficient. Further, it is found that incorporating information on structural breaks in variance improves the accuracy of esti-mating volatility dynamics and effectively reduces the persistence of volatility.