Analysis of detrended time-lagged cross-correlation between two nonstationary time series

Shen, Chenhua

doi:10.1016/j.physleta.2014.12.036

Cited by 92 publications

(30 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The methods discussed are intended for multivariate time series analysis, but they can also be generalized to higher dimensions [38,53,65,66]. We can also use lagged cross-correlations in these methods [67,68]. Although comparing the performances of different methods is always important [69], different variants of a method can produce different outcomes when applied to different systems.…”

Section: Discussionmentioning

confidence: 99%

Detrended partial cross-correlation analysis of two nonstationary time series influenced by common external forces

et al. 2015

View full text Add to dashboard Cite

When common factors strongly influence two power-law cross-correlated time series recorded in complex natural or social systems, using classic detrended cross-correlation analysis (DCCA) without considering these common factors will bias the results. We use detrended partial cross-correlation analysis (DPXA) to uncover the intrinsic power-law cross-correlations between two simultaneously recorded time series in the presence of nonstationarity after removing the effects of other time series acting as common forces. The DPXA method is a generalization of the detrended cross-correlation analysis that takes into account partial correlation analysis. We demonstrate the method by using bivariate fractional Brownian motions contaminated with a fractional Brownian motion. We find that the DPXA is able to recover the analytical cross Hurst indices, and thus the multi-scale DPXA coefficients are a viable alternative to the conventional cross-correlation coefficient. We demonstrate the advantage of the DPXA coefficients over the DCCA coefficients by analyzing contaminated bivariate fractional Brownian motions. We calculate the DPXA coefficients and use them to extract the intrinsic cross-correlation between crude oil and gold futures by taking into consideration the impact of the US dollar index. We develop the multifractal DPXA (MF-DPXA) method in order to generalize the DPXA method and investigate multifractal time series. We analyze multifractal binomial measures masked with strong white noises and find that the MF-DPXA method quantifies the hidden multifractal nature while the MF-DCCA method fails.

show abstract

Section: Discussionmentioning

confidence: 99%

Detrended partial cross-correlation analysis of two nonstationary time series influenced by common external forces

et al. 2015

View full text Add to dashboard Cite

show abstract

“…However, the majority of the existing work focused mainly on the synchronous time series while only limited work considered the cross-correlations between two series with time lags 34 – 37 . In addition, the limited existing work in 34 – 36 , was established on the basis of the detrended fluctuation analysis (DFA) 38 , 39 , and DCCA 14 , 15 . The so-called time-lagged DCCA cross-correlation coefficient was proposed by Shen et al .…”

Section: Introductionmentioning

confidence: 99%

“…The so-called time-lagged DCCA cross-correlation coefficient was proposed by Shen et al . 36 , which can be used to detect time-dependent cross-correlations between the API and wind speed but fails to recognize the range of fluctuation amplitudes that contributed to those cross-correlations. The q -dependent detrended cross-correlation coefficient proposed by Kwapień et al .…”

Section: Introductionmentioning

confidence: 99%

Detecting PM2.5’s Correlations between Neighboring Cities Using a Time-Lagged Cross-Correlation Coefficient

Wang

Chen

2017

Sci Rep

View full text Add to dashboard Cite

In order to investigate the time-dependent cross-correlations of fine particulate (PM2.5) series among neighboring cities in Northern China, in this paper, we propose a new cross-correlation coefficient, the time-lagged q-L dependent height crosscorrelation coefficient (denoted by p q(τ, L)), which incorporates the time-lag factor and the fluctuation amplitude information into the analogous height cross-correlation analysis coefficient. Numerical tests are performed to illustrate that the newly proposed coefficient ρ q(τ, L) can be used to detect cross-correlations between two series with time lags and to identify different range of fluctuations at which two series possess cross-correlations. Applying the new coefficient to analyze the time-dependent cross-correlations of PM2.5 series between Beijing and the three neighboring cities of Tianjin, Zhangjiakou, and Baoding, we find that time lags between the PM2.5 series with larger fluctuations are longer than those between PM2.5 series withsmaller fluctuations. Our analysis also shows that cross-correlations between the PM2.5 series of two neighboring cities are significant and the time lags between two PM2.5 series of neighboring cities are significantly non-zero. These findings providenew scientific support on the view that air pollution in neighboring cities can affect one another not simultaneously but with a time lag.

show abstract

“…The Rozelle-Campbell baseline is "the level of correlation to be expected on the basis of the autocorrelations and synchronous correlations alone" (Lopez-Escobar, et al, 1998:322). Cross-lagged correlations are a developed technique within the agendasetting studies for exploring the similarity between X and Y of causal relationships using time series correlational data shown in Figure 2 (Shen, 2015;Güntert & Wehner, 2015;Latvala, et al 2014;Dunn, 2009;Sweetser et al, 2008;Kinnunen, et al 2008;Lee et al, 2005;Burns, et al 2003;Kivimäki, et al 2000).…”

Section: Results Of Cross-lagged Correlationsmentioning

confidence: 99%

First and Second Level Intermedia Agenda-Setting between International Newspapers and Twitter during the Coverage of the 266th Papal Election

Bakan¹,

Melek²

2018

Akdeniz Üniversitesi İletişim Fakültesi Dergisi

View full text Add to dashboard Cite

This article's main purpose is to understand whether the intermedia agenda-setting effect occurs between the online versions of international newspapers and Twitter during the Papal election in 2013. The researchers have tracked each available country's trending topics on Twitter to understand the popularity of the subject in various countries. The researchers then followed each country available on Twitter at the time when the study was conducted, starting right after the announcement of the new Pope at the Sistine Chapel in Vatican on 12 March 2013 till 15 March 2013 for four days and collected all available data. Later, the researchers collected data from the websites of international newspapers, The New York Times, The Daily Mail, The Guardian, The Telegraph and The Wall Street Journal. Two computer-based content analysis were conducted. In order to understand the relationship amongst these media entities, the cross-lagged panel design with the Rozelle-Campbell (1969) baseline was used. In the end, researchers examined the second level agenda setting effect between media by looking into what was said about the Papal election-positive, negative, or neutral. The results supported both first and second-level agenda setting influence between media.

show abstract

Analysis of detrended time-lagged cross-correlation between two nonstationary time series

Cited by 92 publications

References 27 publications

Detrended partial cross-correlation analysis of two nonstationary time series influenced by common external forces

Detrended partial cross-correlation analysis of two nonstationary time series influenced by common external forces

Detecting PM2.5’s Correlations between Neighboring Cities Using a Time-Lagged Cross-Correlation Coefficient

First and Second Level Intermedia Agenda-Setting between International Newspapers and Twitter during the Coverage of the 266th Papal Election

Contact Info

Product

Resources

About