In recent years, the tendency of the number of financial institutions to include cryptocurrencies in their portfolios has accelerated. Cryptocurrencies are the first pure digital assets to be included by asset managers. Although they have some commonalities with more traditional assets, they have their own separate nature and their behaviour as an asset is still in the process of being understood. It is therefore important to summarise existing research papers and results on cryptocurrency trading, including available trading platforms, trading signals, trading strategy research and risk management. This paper provides a comprehensive survey of cryptocurrency trading research, by covering 146 research papers on various aspects of cryptocurrency trading (e.g., cryptocurrency trading systems, bubble and extreme condition, prediction of volatility and return, crypto-assets portfolio construction and crypto-assets, technical trading and others). This paper also analyses datasets, research trends and distribution among research objects (contents/properties) and technologies, concluding with some promising opportunities that remain open in cryptocurrency trading.
Cryptocurrencies have experienced broad market acceptance and fast development despite their recent conception. Many hedge funds and asset managers have begun to include cryptocurrency-related assets into their portfolios and trading strategies. The academic community has similarly spent considerable efforts in researching cryptocurrency trading. This paper seeks to provide a comprehensive survey of the research on cryptocurrency trading, by which we mean any study aimed at facilitating and building strategies to trade cryptocurrencies.
As an emerging market and research direction, cryptocurrencies and cryptocurrency trading have seen considerable progress and a notable upturn in interest and activity (Farell 2015). From Fig. 1, we observe over 85% of papers have appeared since 2018, demonstrating the emergence of cryptocurrency trading as a new research area in financial trading. The sampling interval of this survey is from 2013 to June 2021.
In this survey we aim at compiling the most relevant research in these areas and extract a set of descriptive indicators that can give an idea of the level of maturity research in this area has achieved.
We also summarise research distribution (among research properties and categories/research technologies). The distribution among properties defines the classification of research objectives and content. The distribution among technologies defines the classification of methods or technological approaches to the study of cryptocurrency trading. Specifically, we subdivide research distribution among categories/technologies into statistical methods and machine learning technologies. Moreover, We identify datasets and opportunities (potential research directions) that have appeared in the cryptocurrency trading area. To ensure that our survey is self-contained, we aim to provide sufficient material to adequately guide financial trading researchers who are interested in cryptocurrency trading.
There has been related work that discussed or partially surveyed the literature related to cryptocurrency trading. Kyriazis (2019) investigated the efficiency and profitable trading opportunities in the cryptocurrency market. Ahamad et al. (2013) and Sharma et al. (2017) gave a brief survey on cryptocurrencies, merits of cryptocurrencies compared to fiat currencies and compared different cryptocurrencies that are proposed in the literature. Mukhopadhyay et al. (2016) gave a brief survey of cryptocurrency systems. Merediz-Sol and Bariviera (2019) performed a bibliometric analysis of bitcoin literature. The outcomes of this related work focused on specific area in cryptocurrency, including cryptocurrencies and cryptocurrency market introduction, cryptocurrency systems / platforms, bitcoin literature review, etc. To the best of our knowledge, no previous work has provided a comprehensive survey particularly focused on cryptocurrency trading.
Definition This paper defines cryptocurrency trading and categorises it into: cryptocurrency markets, cryptocurrency trading models, and cryptocurrency trading strategies. The core content of this survey is trading strategies for cryptocurrencies while we cover all aspects of it.
Multidisciplinary survey The paper provides a comprehensive survey of 146 cryptocurrency trading papers, across different academic disciplines such as finance and economics, artificial intelligence and computer science. Some papers may cover multiple aspects and will be surveyed for each category.
Blockchain is a digital ledger of economic transactions that can be used to record not just financial transactions, but any object with an intrinsic value (Tapscott and Tapscott 2016). In its simplest form, a Blockchain is a series of immutable data records with timestamps, which are managed by a cluster of machines that do not belong to any single entity. Each of these data blocks is protected by cryptographic principle and bound to each other in a chain (cf. Fig. 3 for the workflow).
Confirmation is a critical concept in cryptocurrencies; only miners can confirm transactions. Miners add blocks to the Blockchain; they retrieve transactions in the previous block and combine it with the hash of the preceding block to obtain its hash, and then store the derived hash into the current block. Miners in Blockchain accept transactions, mark them as legitimate and broadcast them across the network. After the miner confirms the transaction, each node must add it to its database. In layman terms, it has become part of the Blockchain and miners undertake this work to obtain cryptocurrency tokens, such as Bitcoin. In contrast to Blockchain, cryptocurrencies are related to the use of tokens based on distributed ledger technology. Any transaction involving purchase, sale, investment, etc. involves a Blockchain native token or sub-token. Blockchain is a platform that drives cryptocurrency and is a technology that acts as a distributed ledger for the network. The network creates a means of transaction and enables the transfer of value and information. Cryptocurrencies are the tokens used in these networks to send value and pay for these transactions. They can be thought of as tools on the Blockchain, and in some cases can also function as resources or utilities. In other instances, they are used to digitise the value of assets. In summary, cryptocurrencies are part of an ecosystem based on Blockchain technology.
Cryptocurrency is a decentralised medium of exchange which uses cryptographic functions to conduct financial transactions (Doran 2014). Cryptocurrencies leverage the Blockchain technology to gain decentralisation, transparency, and immutability (Meunier 2018). In the above, we have discussed how Blockchain technology is implemented for cryptocurrencies.
As of December 20, 2019, there exist 4950 cryptocurrencies and 20,325 cryptocurrency markets; the market cap is around 190 billion dollars (CoinMaketCap 2019). Figure 4 shows historical data on global market capitalisation and 24-h trading volume (TradingView 2021). The blue line is the total cryptocurrency market capitalization and green/red histogram is the total cryptocurrency market volume. The total market cap is calculated by aggregating the dollar market cap of all cryptocurrencies. From the figure, we can observe how cryptocurrencies experience exponential growth in 2017 and a large bubble burst in early 2018. In the wake of the pandemic, cryptocurrencies raised dramatically in value in 2020. In 2021, the market value of cryptocurrencies has been very volatile but consistently at historically high levels.
A cryptocurrency exchange or digital currency exchange (DCE) is a business that allows customers to trade cryptocurrencies. Cryptocurrency exchanges can be market makers, usually using the bid-ask spread as a commission for services, or as a matching platform, by simply charging fees. A cryptocurrency exchange or digital currency exchange (DCE) is a place that allows customers to trade cryptocurrencies. Cryptocurrency exchanges can be market makers (usually using the bid-ask spread as a commission for services) or a matching platform (simply charging fees).
Drastic fluctuations The volatility of cryptocurrencies are often likely to attract speculative interest and investors. The rapid fluctuations of intraday prices can provide traders with great money-earning opportunities, but it also includes more risk.
24-h market The cryptocurrency market is available 24 h a day, 7 days a week because it is a decentralised market. Unlike buying and selling stocks and commodities, the cryptocurrency market is not traded physically from a single location. Cryptocurrency transactions can take place between individuals, in different venues across the world.
Peer-to-peer transactions One of the biggest benefits of cryptocurrencies is that they do not involve financial institution intermediaries. As mentioned above, this can reduce transaction costs. Moreover, this feature might appeal to users who distrust traditional systems.
Scalability problem Before the massive expansion of the technology infrastructure, the number of transactions and the speed of transactions cannot compete with traditional currency trading. Scalability issues led to a multi-day trading backlog in March 2020, affecting traders looking to move cryptocurrencies from their personal wallets to exchanges (Forbes 2021).
Cybersecurity issues As a digital technology, cryptocurrencies are subject to cyber security breaches and can fall into the hands of hackers. Recently, over $600 million of ethereum and other cryptocurrencies were stolen in August 2021 in blockchain-based platform Poly Network (Forbes 2021). Mitigating this situation requires ongoing maintenance of the security infrastructure and the use of enhanced cyber security measures that go beyond those used in traditional banking (Kou et al. 2021).
7fc3f7cf58