Łukasz Czarnecki

Description of multifractality with the MF-DFA method

I will try to show what are the most basic manifestations of mulifractality while analyzing several artificial series with the MF-DFA (Multifractal detrended fluctuation analysis) method. After a short description of multifractality itself, I will present several methods of generating artificial data with different kinds of fractality. This will be followed by a discussion of MF-DFA characteristics calculated for the fore mentioned series. Finally, I will show the impact of small (manually introduced) nonstationarity on the results and a comparison between MF-DFA and its alternative: WTMM (wavelets transform modulus maxima).