File Exchange

image thumbnail

r-DFA : Robust Detrended Fluctuation Analysis

version (3.54 MB) by Abrar Habib
Robust Detrended Fluctuation Analysis


Updated 30 Apr 2017

View License

This MATLAB package is used to perform Robust Detrended Fluctuation Analysis (herein referred to as r-DFA). r-DFA is a procedure that feeds the results of the regular DFA to a series of statistical models in order to:
- determine a robust estimate of the global scaling exponent using Robust Regression,
- and to determine crossovers that produce statistically significant scaling exponents using Piecewise Linear Regression (a modification of Guido Albertin’s – 2013 PLR.m), ANCOVA and Multiple Comparison Procedure.
CITING: Please cite the following paper when using this code:
Abrar Habib, James P.R. Sorensen, John P. Bloomfield, Katie Muchan, Andrew J. Newell, Adrian P. Butler, Temporal scaling phenomena in groundwater-floodplain systems using robust detrended fluctuation analysis, Journal of Hydrlogy (2017), 549, pp. 715-730

A comprehensive explanation is available in the ReadMe.txt and a detailed explanation is available in the aforementioned paper.

Comments and Ratings (5)

many thanks for the helpful code

rui zhang

many many many thanks to you


Thanks for sharing

A systematic robust method to use detrended Fluctuation Analysis in the field of water resources. It is really helpful/


Minor changes to description.

Notes for citation of the code have been added to the description.

MATLAB Release Compatibility
Created with R2016b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Inspired by: piecewise.m, Piecewise linear least square fit