Algorithmic Cryptocurrency Trading

Algorithmic trading has become an integral part of the financial markets. However, a lack of publicly available programming interfaces and overly expensive fee structures make it difficult for small investors to implement their own trading algorithms. In the meantime, the emerging asset class of cryptocurrencies offers many possibilities, as most exchanges provide freely available programming interfaces and low trading fees. In the course of this master thesis, I examine whether profitable algorithmic cryptocurrency trading is possible. The approach taken in this research is based on a self-developed algorithmic trading system, consisting of three strategies that invest according to the principles of trend following and technical analysis. The results show that algorithmic cryptocurrency trading can be profitable, as the developed trading system achieves a net return of 27.8% in Euro terms. However, not all strategies win, which shows that an automated investment approach also entails a high risk of loss.