Structure of a financial cross-correlation matrix under attack

Lim, Gyuchang; Kim, SooYong; Kim, Junghwan; Kim, Pyungsoo; Kang, Yoonjong; Park, Sanghoon; Park, Inho; Park, Sang‐Bum; Kim, Kyungsik

doi:10.1016/j.physa.2009.05.018

“…In this paper, we have extended the method of multifractal-detrended-cross-correlation-analysis for the case when more than two parameters are coupled to each other. There are other methods, such as random matrix theory [37][38][39][40][41][42][43] and complex networks [44][45][46][47], which consider the coupling between parameters. But they should be applied on stationary time series.…”

Section: Introductionmentioning

confidence: 99%

Coupling detrended fluctuation analysis for analyzing coupled nonstationary signals

Hedayatifar

¹

,

Vahabi

²

,

Jafari

³

2011

View full text Add to dashboard Cite

When many variables are coupled to each other, a single case study could not give us thorough and precise information. When these time series are stationary, different methods of random matrix analysis and complex networks can be used. But, in nonstationary cases, the multifractal-detrended-cross-correlation-analysis (MF-DXA) method was introduced for just two coupled time series. In this article, we have extended the MF-DXA to the method of coupling detrended fluctuation analysis (CDFA) for the case when more than two series are correlated to each other. Here, we have calculated the multifractal properties of the coupled time series, and by comparing CDFA results of the original series with those of the shuffled and surrogate series, we can estimate the source of multifractality and the extent to which our series are coupled to each other. We illustrate the method by selected examples from air pollution and foreign exchange rates.

show abstract

“…The analysis of financial time-series has attracted the interest of many researchers [17,27,37], as large amounts of data are freely available for study. Moreover, the universality exhibited by financial markets is shared with other CS.…”

Section: Stock Markets (St)mentioning

confidence: 99%

Power Law Behavior and Self-Similarity in Modern Industrial Accidents

Lopes

¹

,

Machado

²

2015

Int. J. Bifurcation Chaos

View full text Add to dashboard Cite

This paper analyzes several natural and man-made complex phenomena in the perspective of dynamical systems. Such phenomena are often characterized by the absence of a characteristic length-scale, long range correlations and persistent memory, which are features also associated to fractional order systems. For each system, the output, interpreted as a manifestation of the system dynamics, is analyzed by means of the Fourier transform. The amplitude spectrum is approximated by a power law function and the parameters are interpreted as an underlying signature of the system dynamics. The complex systems under analysis are then compared in a global perspective in order to unveil and visualize hidden relationships among them.

show abstract

“…ST are CS originated by different types of dynamical processes and characterized by universal PLs. The analysis of financial time-series has attracted the interest of many researchers [17,27,37], as large amounts of data are freely available for study. Moreover, the universality exhibited by financial markets is shared with other CS.…”

Section: Stock Markets (St)mentioning

confidence: 99%

The Persistence of Memory

Machado

¹

,

Lopes

²

2014

View full text Add to dashboard Cite

This paper analyzes several natural and man-made complex phenomena in the perspective of dynamical systems. Such phenomena are often characterized by the absence of a characteristic length-scale, long range correlations and persistent memory, which are features also associated to fractional order systems. For each system, the output, interpreted as a manifestation of the system dynamics, is analyzed by means of the Fourier transform. The amplitude spectrum is approximated by a power law function and the parameters are interpreted as an underlying signature of the system dynamics. The complex systems under analysis are then compared in a global perspective in order to unveil and visualize hidden relationships among them.

show abstract

Structure of a financial cross-correlation matrix under attack

Cited by 20 publications

References 19 publications

Coupling detrended fluctuation analysis for analyzing coupled nonstationary signals

Coupling detrended fluctuation analysis for analyzing coupled nonstationary signals

Power Law Behavior and Self-Similarity in Modern Industrial Accidents

The Persistence of Memory

Contact Info

Product

Resources

About