Cryptocurrency trading has become more and more popular among private investors. According to recent studies, the momentum effect influences the underlying market. Quantitative trading systems can leverage momentum indicators to open and close trading positions. However, existing approaches that exploit the momentum effect in cryptocurrency trading do not rely on machine learning. Since these systems are based on human generated rules they are not suited to highly volatile market conditions, which are quite common in cryptocurrency markets. This paper proposes to leverage machine learning approaches to automatically detect the momentum effect in cryptocurrency market data. For each cryptocurrency it estimates the likelihood of being affected by the momentum effect on the next trading day as well as the momentum direction. A backtesting session, performed on three very popular cryptocurrencies, shows that the machine learning models are able to predict, to a good approximation, short-term price volatility thus reducing the number of false trading signals and increasing the return on investments compared to state-of-the-art approaches.
In recent years, cryptocurrencies have gained the attention of financial institutions, and, consequently, the speculative interest in Bitcoin, Ethereum, and other cryptocurrencies has significantly increased (Fang et al., 2020). Cryptocurrencies are assets characterized by peculiar price trends and exchange volumes. This is mainly due to the medium of exchange and the ownership policies, which commonly yield a significantly higher degree of price volatility compared to conventional assets (King & Koutmos, 2021). This poses the question of whether the underlying market adheres to the predictive models that are commonly applied to more traditional markets such as the stock and Forex exchanges.
The models belonging to the traditional finance theory require investors to be risk-averse and able to make rational choices. The purpose is to maximize profits without being influenced by other factors and have complete access to all the information available in the market. These models also require an effective arbitrage mechanism, which plays a critical role in determining the prices of the securities (Sinkala, 2016). The arbitrage mechanism entails giving investors the opportunity to earn by buying/selling the same asset at a lower/higher price respectively, When an arbitrage opportunity arises, it is essential that this opportunity is immediately exploited by investors. In this way the market will allow the prices to return to the right equilibrium immediately, not allowing an asset to be overvalued or undervalued for too long periods. Notably, the aforesaid assumptions are partly unrealistic for the cryptocurrency market (Fang et al., 2020). These reasons are the origin of anomalous phenomena observed in cryptocurrency markets, i.e., the momentum and reversal effects.
6. Conclusions and future work
The paper investigated the use of machine learning techniques to overcome the limitations of state-of-the-art momentum-based cryptocurrency trading systems. Specifically, based on the empirical observation that the momentum effect is likely to influence the series of cryptocurrency prices, we designed a methodology that predicts the presence and direction of an overreaction condition. The takeaways of the research are summarized below:
• The return per trade of the machine learning-based approach is significantly better than those achieved by the heuristic approach (e.g., on BTC KNN 0.78% vs. Heuristic approach 0.06%, on ETH KNN 1.87% vs. Heuristic approach 0.71%).