Abstract
In this paper, we examine the dynamic cross‐correlations between participants’ attentions to the P2P lending and offline loan (lending) with the method of multifractal detrended cross‐correlation analysis (MF‐DCCA). The empirical result mainly shows that (1) the power‐law cross‐correlation exists between participants’ attentions to the P2P lending and offline loan and is persistent, (2) the cross‐correlation is more stable in the short term, and (3) the relation subjected to a small fluctuation is more cross‐correlated than that under larger ones. Furthermore, we carry out the robustness test to verify the result. The Granger causality test indicates that participants’ attentions to P2P lending and offline loan Granger cause each other in the short term.
Highlights
E Pearson, Spearman, and Kendall correlations are conventional methods extensively used in investigating the relationship between two variables. e Pearson coefficient measures linear correlation, yet its application relies on the precondition that the time series are stationary and obey the normal distribution
While multifractality is ubiquitously observed in socioeconomic systems [36], the aforementioned traditional correlation methodologies are not tested and verified by existing documents to examine multifractality. erefore, both detrended crosscorrelation analysis (DCCA) advanced by Podobnik and Stanley [37] and multifractal detrended cross-correlation analysis (MF-DCCA) proposed by Zhou [38] are selected to study the cross-correlations between participants’ attentions to P2P lending and offline loan
We focus on the private lending network system which consists of P2P lending and offline loan and investigate the cross-correlations of participants’ attention between two types of lending
Summary
E Pearson, Spearman, and Kendall correlations are conventional methods extensively used in investigating the relationship between two variables. e Pearson coefficient measures linear correlation, yet its application relies on the precondition that the time series are stationary and obey the normal distribution. Kristoufek [40] confirms that, for the nonstationary series, the DCCA coefficient measures correlation accurately despite various levels of nonstationarity and dominates the Pearson coefficient. Compared with the Pearson correlation coefficients, the DCCA coefficients are scale-dependent and more robust to the amplitude ratio between slow and fast components and contaminated noises [45]. We obtain daily search volume data of P2P lending and offline loan from Baidu Search Index. E P2P lending search volume reflects participants’ attention. E offline loan search volume has a similar meaning. The search volumes can represent participants’ attention to the approach of borrowing or lending in the private lending network. In order to explore cross-correlations between participants’ attention to P2P lending and that to offline loan, we define two variables as follows: P2Pt ln P2Pt,. Loant ln loant, where P2Pt is the Baidu index of P2P lending and loant is the Baidu index of offline loan in time t, respectively
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
