Fractionally integrated process for transition economics

We analyze the European transition economics and show that many time series of major indices exhibit (i) power-law correlations in their values, (ii) power-law correlations in their magnitudes and (iii) an asymmetric probability distribution. Applying the phase randomization procedure to these time series, we show that magnitude correlations completely vanish. We propose a stochastic model that can generate time series with features (i), (ii) and (iii), and we show by means of numerical simulations that this model is capable of reproducing these three features found in the empirical data. (c) 2005 Elsevier B.V. All rights reserved.