Title: Empirical comparison of preferential attachment and linking statistics in Bitcoin and Ethereum
Authors: Daniel Kondor - SMART (Singapore) [presenting]
Gabor Vattay - ELTE (Hungary)
Istvan Csabai - ELTE (Hungary)
Jozsef Steger - ELTE (Hungary)
Nikola Bulatovic - ELTE (Hungary)
Abstract: Cryptocurrencies have presented a disruptive change for both economics and computer science. Considering the list of transactions as an evolving network, cryptocurrencies are among the largest real-world networks that can be analyzed by the scientific community, with several hundred million total edges. While there is significant interest in how cryptocurrencies work from a network science perspective, we still do not have a comprehensive understanding of which are the relevant processes that shape the network structure. We evaluate key network characteristics on the Bitcoin and Ethereum transaction networks, the two most popular cryptocurrencies. We specifically look at network evolution and the dynamics of how nodes gain new transaction partners and gain or lose balance. We show that a process of preferential attachment continues to be determinant for both cryptocurrencies and is robust concerning the time period analysed and the method used to reconstruct the transaction network. During our analysis, we perform an in-depth comparison among Bitcoin and Ethereum, focusing on comparing the transaction dynamics of regular addresses in the two systems and between addresses and smart contracts in Ethereum. In all cases, we evaluate correlations between node degree, activity and wealth.