City academic makes link between cryptocurrency trading and machine learning
A new study led by Dr Andrea Baronchelli, published on arXiv, has tested the connection between machine learning and artificial intelligence-assisted trading and the cryptocurrency market.
Dr Baronchelli has analysed daily data for 1681 cryptocurrencies between November 2015 and April 2018 and has shown that the inefficiency of this young market can be exploited to generate abnormal profits.
“Our results need to be considered with caution, as we did not consider some of the real-world complications of the market, such as fees, etc”, Dr Baronchelli says about the study, titled, Machine Learning the Cryptocurrency Market.
He remarks, however, that the results demonstrate that "non-trivial, but ultimately simple, algorithmic mechanisms can help anticipate the short-term evolution of the cryptocurrency market”.
This study forms part of a consolidated line of research carried out by Dr Baronchelli’s team at City, complementing earlier results on the long-term stability of the cryptocurrency market.
Alongside his analysis of market behaviour, Dr Baronchelli is also investigating new forms of governance enabled by blockchain, specifically the behaviour of Decentralised Autonomous Organisations (or DAO).
In collaboration with the Dash Intel team, he will be studying the data behind the Dash Treasury and exploring the 'Decentralized Governance by Blockchain' (DGBB) model.
The research will be focused on analyzing how the Dash ecosystem and community has evolved since the introduction of the DGBB, and weighing the benefits and trade-offs of decentralized governance models in relation to traditional organizational structures.