Detrended fluctuation analysis as a regression framework: estimating dependence at different scales.

We propose a framework combining detrended fluctuation analysis with standard regression methodology. The method is built on detrended variances and covariances and it is designed to estimate regression parameters at different scales and under potential nonstationarity and power-law correlations. The former feature allows for distinguishing between effects for a pair of variables from different temporal perspectives. The latter ones make the method a significant improvement over the standard least squares estimation. Theoretical claims are supported by Monte Carlo simulations. The method is then applied on selected examples from physics, finance, environmental science, and epidemiology. For most of the studied cases, the relationship between variables of interest varies strongly across scales.

[1]  Wei‐Xing Zhou.  Multifractal detrended cross-correlation analysis for two nonstationary signals. , 2008, Physical review. E, Statistical, nonlinear, and soft matter physics.

[2]  H E Stanley,et al.  Statistical physics and physiology: monofractal and multifractal approaches. , 1999, Physica A.

[3]  A. Goldberger,et al.  Finite-size effects on long-range correlations: implications for analyzing DNA sequences. , 1993, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[4]  P. Varotsos,et al.  Long-range correlations in the electric signals that precede rupture. , 2002, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  Peter Talkner,et al.  Spectra and correlations of climate data from days to decades , 2001 .

[6]  H. Stanley,et al.  Cross-correlations between volume change and price change , 2009, Proceedings of the National Academy of Sciences.

[7]  S. Havlin,et al.  Detecting long-range correlations with detrended fluctuation analysis , 2001, cond-mat/0102214.

[8]  S. Havlin,et al.  Correlated and uncorrelated regions in heart-rate fluctuations during sleep. , 2000, Physical review letters.

[9]  Zhi-Qiang Jiang,et al.  Multifractal detrending moving-average cross-correlation analysis. , 2011, Physical review. E, Statistical, nonlinear, and soft matter physics.

[10]  P. Cizeau,et al.  CORRELATIONS IN ECONOMIC TIME SERIES , 1997, cond-mat/9706021.

[11]  G. F. Zebende.  DCCA cross-correlation coefficient: Quantifying level of cross-correlation , 2011 .

[12]  H. Stanley,et al.  Quantification of scaling exponents and crossover phenomena in nonstationary heartbeat time series. , 1995, Chaos.

[13]  H. Stanley,et al.  Multifractal Detrended Fluctuation Analysis of Nonstationary Time Series , 2002, physics/0202070.

[14]  H. Stanley,et al.  Analysis of DNA sequences using methods of statistical physics , 1998 .

[15]  R. Mantegna,et al.  Long-range correlation properties of coding and noncoding DNA sequences: GenBank analysis. , 1995, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[16]  Talkner,et al.  Power spectrum and detrended fluctuation analysis: application to daily temperatures , 2000, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[17]  Arjen van Ooyen,et al.  Altered temporal correlations in parietal alpha and prefrontal theta oscillations in early-stage Alzheimer disease , 2009, Proceedings of the National Academy of Sciences.

[18]  H E Stanley,et al.  Statistical properties of DNA sequences. , 1995, Physica A.

[19]  Ladislav Kristoufek,et al.  Measuring correlations between non-stationary series with DCCA coefficient , 2013, 1310.3984.

[20]  C. Peng,et al.  Mosaic organization of DNA nucleotides. , 1994, Physical review. E, Statistical physics, plasmas, fluids, and related interdisciplinary topics.

[21]  H. Stanley,et al.  Detrended cross-correlation analysis: a new method for analyzing two nonstationary time series. , 2007, Physical review letters.

[22]  Tobias Preis,et al.  Adaptive nowcasting of influenza outbreaks using Google searches , 2014, Royal Society Open Science.