Publication detail

Kristoufek, L.: Detrended fluctuation analysis as a regression framework: Estimating dependence at different scales

Author(s): prof. PhDr. Ladislav Krištoufek Ph.D.,
Type: Articles in journals with impact factor
Year: 2015
Number: 0
ISSN / ISBN:
Published in: Physical Review E 91, art. 022802 arXiv PDF
Publishing place:
Keywords: detrended fluctuation analysis, regression, scales, time series analysis
JEL codes:
Suggested Citation:
Grants: GACR P402/11/0948 Developing Analytical Framework for Energy Security: Time-Series Econometrics, Game Theory, Meta-Analysis and Theory of Regulation GAČR 14-11402P Bivariate long memory analysis of financial time series (2014-2016)
Abstract: 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 non-stationarity 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.

Partners

Deloitte

Sponsors

CRIF
McKinsey
Patria Finance