Tensor Change-Point Detection Across Cryptocurrency Trading Networks

Summary

The paper introduces TenSeg, a method for detecting structural changes over time across multiple related transaction networks. It addresses a challenge in fraud and market-manipulation analysis: activity may span several cryptocurrency platforms, so a shift can be difficult to identify by studying one network alone. TenSeg decomposes tensor-valued observations and then searches for multiple change points in the second-order dependence structure of the resulting data, corresponding to changes in cross-covariance or network structure.

The authors evaluate the approach on simulated datasets and Ethereum blockchain network data. They report that it outperforms other contemporary change-point methods in their experiments, and that detected Ethereum changes align with shifts across multiple trading networks. The method is designed to handle frequent changes that may be small in magnitude and is described as computationally fast. The available description does not provide the experimental design, performance measures, or details needed to assess robustness beyond those datasets, so its reported advantage should be read as evidence from the specific simulations and Ethereum application.

Key ideas

• TenSeg first decomposes tensor-valued network observations and then detects changes in their second-order dependence structure.
• The method is designed to locate multiple changes, including frequent changes with potentially small magnitudes.
• Studying several connected trading networks can help identify manipulation that spans multiple cryptocurrency platforms.
• The paper evaluates TenSeg on simulated data and Ethereum blockchain network data.
• The reported results are promising, but the description does not specify evaluation metrics or broader robustness tests.

Tags

# Tensor time series change-point detection in cryptocurrency network data


# Tensor time series change-point detection in cryptocurrency network data









Financial fraud has been growing exponentially in recent years. The rise of cryptocurrencies as an investment asset has simultaneously seen a parallel growth in cryptocurrency scams. To detect possible cryptocurrency fraud, and in particular market manipulation, previous research focused on the detection of changes in the network of trades; however, market manipulators are now trading across multiple cryptocurrency platforms, making their detection more difficult. Hence, it is important to consider the identification of changes across several trading networks or a `network of networks' over time. To this end, in this article, we propose a new change-point detection method in the network structure of tensor-variate data. This new method, labeled TenSeg, first employs a tensor decomposition, and second detects multiple change-points in the second-order (cross-covariance or network) structure of the decomposed data. It allows for change-point detection in the presence of frequent changes of possibly small magnitudes and is computationally fast. We apply our method to several simulated datasets and to a cryptocurrency dataset, which consists of network tensor-variate data from the Ethereum blockchain. We demonstrate that our approach substantially outperforms other state-of-the-art change-point techniques, and the detected change-points in the Ethereum data set coincide with changes across several trading networks or a `network of networks' over time. Finally, all the relevant \textsf{R} code implementing the method in the article are available on https://github.com/Anastasiou-Andreas/TenSeg.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.