MULTIFRACTAL ANALYSIS OFTHE DYNAMICS OF TURKISHEXCHANGE RATE

We perform a comparative study of applicability of the Multifractal DetrendedFluctuation Analysis (MFDFA) and the Wavelet Transform Modulus Maxima(WTMM) method in properly detecting ofmono- and multifractal character ofdata. After summarizing the theory behind both methods, we apply both methodson USD/TRY currency. The results show thatour data has multifractal nature butnot at high level and multifractality ispoorer if WTMM method is used. We alsoinvestigated whether other Eastern European country currencies, such as RussianRubble and Hungarian Forint have multifractal characters by using MFDFAmethod. Therefore, forecasters have often encountered in trying to predict theseexchange rates with models that do notincorporate any notion of inhomogeneitywill have little predictive power