Probabilistic properties of detrended fluctuation analysis for Gaussian processes

Abstract: The detrended fluctuation analysis (DFA) is one of the most widely used tools for the detection of long-range correlations in time series. Although DFA has found many interesting applications and has been shown as one of the best performing detrending methods, its probabilistic foundations are still unclear. In this paper we study probabilistic properties of DFA for Gaussian processes. The main attention is paid to the distribution of the squared error sum of the detrended process. This allows us to find the expected value and the variance of the fluctuation function of DFA for a Gaussian process of general form. The results obtained can serve as a starting point for analyzing the statistical properties of the DFA-based estimators for the fluctuation and correlation parameters. The obtained theoretical formulas are supported by numerical simulations of particular Gaussian processes possessing short-and long-memory behaviour.