This is a simple Python function (ab_filter) that calculates
continuous variation in scaling exponents by applying alpha-beta
filter to the DFA results. This procedure was reported in Echeverria
et al. "Interpretation of heart rate variability via detrended
fluctuation analysis and ab filter", Chaos 2003; 13(2):467-75.
The main output of a DFA routine is window sizes and fluctuation levels. When plotted in log-log they show linear tendencies, which are summarized by a scaling exponent. However, sometimes one might encounter phenomena that deviate from truly linear relationship, for example, cross-over is to be described by two linear fits and two exponents.
Alpha-Beta filter takes as an input log-fluctuations and log-window sizes to produce a continous plot of the scaling exponent variation over window sizes.
Consult the help of the function ab_filter for further details on
usage and see information on alpha-beta filter elsewhere.
A single principal function ab_filter constitutes the package (with
two ancillary functions). Dependency: python3 + NumPy.