Kyoto University researchers from Japan published a study about XRP price and transaction network structure, with a strong anti-correlation during bubble periods.
Graduate researchers at Kyoto University in Japan have recently published a report capturing the dynamic relationship between the price of XRP and the correlation tensor spectra of its transaction network.
For context, the correlation tensor spectra of a network is a mathematical representation of the network’s structure and connectivity. It is to identify patterns and relationships between different nodes.
Prominent XRP influencer WrathofKahneman (WoK) spotlighted the research report to the XRP community in a recent tweet.
Recent study out of Kyoto University examines #XRP price during/after '18 bubble and correlations. Most interesting? (2021 is latest data) They identify a set of driver nodes during the bubble. Supported by #Ripple Impact Fund.https://t.co/K8o0lvSBNp pic.twitter.com/tdAxamRtZL
— WrathofKahneman 🪝 (@WKahneman) September 22, 2023
The Research Report
The researchers elaborated on a previous study spanning the periods around 2018, commonly dubbed the “crypto bubble.” That study found that the price of XRP exhibited a vital anti-correlation with “the largest singular values derived from the correlation tensors extracted from weekly XRP transaction networks.”
In simpler terms, it found a consistent and strong pattern: when the price of XRP went up, specific significant values calculated from the transaction data went down, and when the price of XRP went down, these values went up.
Notably, this follow-up research extended the methodology of correlation tensor spectra for XRP transaction networks. The researchers estimated and compared the “large singular values” distribution.
In their analysis, they employed “random matrix theory and empirical correlation tensor data.” Notably, their investigation spanned two years, encompassing both the bubble and non-bubble phases of the XRP market.
In non-bubble periods, there was no significant correlation between XRP’s price and the largest singular value. However, the research detected a strong anti-correlation between XRP’s price and the largest singular value during the bubble period.
Using the information derived from the singular vectors, the researchers also pinpointed a set of driver nodes that played a pivotal role in influencing the XRP market during the bubble period.
Ultimately, the study showcased the potential of correlation tensor spectra in understanding the dynamic interplay between XRP’s price and transaction network. It highlighted how market conditions influence this relationship and provide insights into the mechanics of the XRP market during bubble and non-bubble periods.