| Issue |
EPL
Volume 81, Number 1, January 2008
|
|
|---|---|---|
| Article Number | 18002 | |
| Number of page(s) | 4 | |
| Section | Interdisciplinary Physics and Related Areas of Science and Technology | |
| DOI | https://doi.org/10.1209/0295-5075/81/18002 | |
| Published online | 03 December 2007 | |
Multifractality and heteroscedastic dynamics: An application to time series analysis
1
Instituto de Física, Universidade Federal de Alagoas - 57072-970, Maceió-AL, Brazil
2
Instituto de Ciências Humanas, Comunicação e Artes, Universidade Federal de Alagoas 57072-970, Maceió-AL, Brazil
3
Dipartimento di Matematica and INFM, Università dell'Aquila - I-67010 L'Aquila, Italy
Received:
26
July
2007
Accepted:
30
October
2007
An increasingly important problem in physics concerns scale invariance symmetry in diverse complex systems, often characterized by heteroscedastic dynamics. We investigate the nature of the relationship between the heteroscedastic and fractal aspects of the dynamics of complex systems, by analyzing the sensitivity to heteroscedasticity of the scaling properties of weakly nonstationary time series. By using multifractal detrended fluctuation analysis, we study the singularity spectra of currency exchange rate fluctuations, after partially or completely eliminating n-point correlations via data shuffling techniques. We conclude that heteroscedasticity can significantly increase multifractality and interpret these findings in the context of self-organizing and adaptive complex systems.
PACS: 89.65.Gh – Economics; econophysics, financial markets, business and management / 89.20.-a – Interdisciplinary applications of physics / 02.50.-r – Probability theory, stochastic processes, and statistics
© EPLA, 2008
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.