Box counting is commonly used as a method to determine the multifractal spectrum of measured time series. Here we show that delta-correlated signals, having an (hyper)exponential amplitude distribution, can generate a spurious detection of multifractality. The situation becomes even worse for nonlinearly-filtered, Linear autoregressive processes. Care should thus be exerted when claiming the presence of multifractality in measured data.
A box-counting red herring
von Hardenberg J;Provenzale;
2000
Abstract
Box counting is commonly used as a method to determine the multifractal spectrum of measured time series. Here we show that delta-correlated signals, having an (hyper)exponential amplitude distribution, can generate a spurious detection of multifractality. The situation becomes even worse for nonlinearly-filtered, Linear autoregressive processes. Care should thus be exerted when claiming the presence of multifractality in measured data.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
