Affiliations: [a] Georgian Technical University, Tbilisi, Georgia | [b] Ivane Javakhishvili State University, Tbilisi, Georgia
Correspondence:
[*]
Corresponding author: Alexander L. Topchishvili, Friedrich-Ebert-Str. 88, 35039 Marburg, Germany. E-mail: alextopchi@gmail.com.
Abstract: The paper has been devoted to the investigation of statistical, distributional and dynamical features of exchange rate variation of Georgian currency (Georgial Lari – GEL) to the US Dollar within the last 15 years, since 1996 (when GEL was put into operation) upto the end of 2010. Dispersion characteristics, third and fourth statistical moments, probability distribution function of all existing currency exchange data sets as well as annual exchange rate time series have been calculated. Probability distribution fitting procedure has been used to fit observed exchange rate data to numerous hypothesized statistical models and the statistical goodness of fit methods were used to assess the quality of fitting and to identify the best candidate distribution for a given scenario. Scaling and dynamical features of exchange rate variation have been assessed using modern linear and nonlinear methods of time series analysis. For this, wavelet coefficients and detrended fluctuation exponents were calculated. Recurrence quantification analysis has been used to quantify extent of order in exchange rate variation. The same scaling and dynamical analysis has been carried out for the whole timeframe and for one year exchange rate data sets as well. Obtained results of exchange rate variation calculations have been tested by shuffled surrogate method. Time scale dependence of national currency exchange rate variation was confirmed for USD/GEL exchange rate data sets. Statistical, linear and nonlinear features of the exchange rate process were essentially changed for different observation time periods. Increased long range correlations and extent of order in the variation of USD/GEL exchange rate was found in time periods when Georgian economy was affected by the strong internal and international crises. This indicates the clear relationship between internal conditions and external influences with the Georgian domestic economic processes. According to the conducted analysis general features of statistical and dynamical changes in processes of small economy are quite similar notwithstanding are they caused by internal development, external influences or by their combination.
Keywords: Statistics, probability distribution, dynamics, scaling, exchange rate
