A Comprehensive Survey on Cryptocurrency Trading

Introduction

In recent years, the integration of cryptocurrencies into investment portfolios by financial institutions has accelerated significantly. As the first pure digital assets adopted by asset managers, cryptocurrencies share some common traits with traditional assets but possess a distinct nature. Their behavior as an asset class is still being extensively studied. This makes it crucial to consolidate existing research and findings on cryptocurrency trading, which encompasses trading platforms, signals, strategy development, and risk management.

This comprehensive survey reviews 146 research papers across various dimensions of cryptocurrency trading, including trading systems, bubble and extreme condition analysis, volatility and return prediction, portfolio construction, and technical trading methodologies. It further analyzes datasets, research trends, and the distribution of research across different domains and technologies, concluding with open opportunities in the field.

Understanding Cryptocurrency Markets

What Are Cryptocurrencies?

Cryptocurrencies are decentralized digital mediums of exchange that utilize cryptographic functions to conduct financial transactions. Leveraging blockchain technology, they achieve decentralization, transparency, and immutability. Their security is underpinned by advanced cryptography rather than trust or intermediaries. For instance, Bitcoin employs Elliptic Curve Cryptography, making transactions exceedingly secure.

Key characteristics include a controlled supply—many cryptocurrencies cap their total volume. Bitcoin’s supply, for example, is designed to decrease over time, reaching its final number around 2140. A pivotal feature is the elimination of financial intermediaries, reducing transaction costs. The decentralized nature of blockchain also makes cryptocurrencies theoretically immune to government interference.

As of December 2019, there were 4,950 cryptocurrencies and 20,325 markets, with a total market capitalization of approximately $190 billion. The market has seen exponential growth, notably in 2017, a major bubble burst in early 2018, and a dramatic rise during the 2020 pandemic.

Major Cryptocurrencies

The three mainstream cryptocurrencies are:

• Bitcoin (BTC): Launched in 2009, it is described as a pure peer-to-peer electronic cash system.
• Ethereum (ETH): Launched in 2015, it features a special token called Ether and allows the creation of new tokens on its blockchain.
• Litecoin (LTC): A peer-to-peer cryptocurrency based on the Bitcoin protocol but using a different hashing algorithm (Scrypt).

Bitcoin and Ethereum dominate the total market capitalization.

Cryptocurrency Exchanges

Cryptocurrency exchanges are platforms that allow customers to trade digital currencies. They can function as market makers, earning through the bid-ask spread, or as matching platforms that charge fees. Prominent exchanges include the Chicago Mercantile Exchange (CME), Binance, and Coinbase, which vary in supported currencies, fiat options, and regulatory oversight.

Cryptocurrency Trading Defined

Cryptocurrency trading involves buying and selling digital assets with the intention of generating a profit. It can be broken down into three aspects:

1. Object: The asset being traded (i.e., the cryptocurrency itself).
2. Operation Mode: The means of transaction, such as trading Contracts for Difference (CFDs) or directly buying/selling on exchanges.
3. Trading Strategy: A set of predefined rules or algorithms guiding buy and sell decisions.

Advantages and Disadvantages

Advantages:

• High Volatility: Offers significant profit opportunities, albeit with higher risk.
• 24/7 Market: Operates round-the-clock due to its decentralized nature.
• Anonymity: Provides privacy in transactions, though exchanges implement Know-Your-Customer (KYC) measures.
• Peer-to-Peer Transactions: Eliminates intermediaries, reducing costs.
• Programmability: Some cryptocurrencies offer additional utilities like voting rights or partial ownership of physical assets.

Disadvantages:

• Scalability Issues: Current infrastructure can lead to transaction delays.
• Cybersecurity Risks: Susceptible to breaches and hacking.
• Regulatory Uncertainty: Varies globally, with some countries adopting strict measures while others remain cautious.

Trading Strategies and Systems

Cryptocurrency Trading Strategies

Trading strategies are broadly categorized into technical and fundamental analysis. Technical analysis relies on historical data and patterns, while fundamental analysis evaluates intrinsic value based on economic indicators. A third category, programmatic trading, uses quantitative data and algorithms to exploit market inefficiencies automatically.

Software Trading Systems

Several cryptocurrency trading systems facilitate automated trading:

• Capfolio: A professional analysis platform with an advanced backtesting engine.
• 3Commas: Supports simultaneous take-profit and stop-loss orders.
• CCXT: A library with a unified API for multiple exchanges.
• Blackbird: Executes market-neutral arbitrage strategies.
• Freqtrade: A Python-based bot with machine learning optimization.
• CryptoSignal: Focuses on technical analysis and alerts.

These systems help overcome challenges like price manipulation and transaction delays.

Systematic Trading

Systematic trading involves predefined rules for entry, exit, and risk management. It includes:

• Technical Analysis: Uses historical data and indicators like moving averages and RSI to predict future movements.
• Pairs Trading: Exploits price discrepancies between two correlated assets.
• Informed Trading: Leverages non-public information for advantage, evidenced in Bitcoin markets.

Emerging Technologies

Econometrics: Applies statistical models to economic data. Common methods include:

• GARCH Models: Analyze volatility clustering.
• BEKK Models: Examine shock transmission and spillover effects.
• Granger-Causality Tests: Assess causal relationships between variables like trading volume and returns.

Machine Learning: Enhances prediction accuracy by identifying complex patterns. Approaches include:

• Supervised Learning: For classification and regression tasks (e.g., predicting price directions).
• Unsupervised Learning: For clustering and pattern recognition.
• Reinforcement Learning: Trains agents to maximize cumulative rewards through interaction.

Common algorithms include Random Forests, SVMs, LSTMs, and CNNs. Studies show that models like LSTMs perform well in capturing long-term dependencies in high-volatility environments.

👉 Explore advanced trading algorithms

Portfolio Management and Market Conditions

Portfolio Theory

Diversification is key to managing risk and maximizing returns. In cryptocurrency, this can involve:

• Diversifying Across Cryptocurrencies: Spreading investments across multiple digital assets.
• Global Market Diversification: Including traditional assets like stocks and futures.

Market Conditions

Cryptocurrency markets are prone to bubbles and extreme volatility. The 2017 boom and subsequent crash highlighted the need for robust risk management. Continuous bull markets, like that in late 2020, further emphasize the importance of understanding market cycles and maintaining exposure limits.

Research Trends and Analysis

Paper Collection and Scope

This survey adopted a bottom-up approach, collecting papers from databases like Google Scholar and arXiv. Keywords included "cryptocurrency trading," "prediction," "risk management," and "portfolio." The snowballing method ensured comprehensive coverage across six research areas.

Distribution and Trends

Research distribution shows:

• Nearly 49% of papers published in finance and economics journals.
• Significant contributions from computer science and AI venues.
• Increasing focus on machine learning and econometrics.

Datasets commonly used include historical price data, social media sentiment, and technical indicators.

Frequently Asked Questions

What is cryptocurrency trading?
Cryptocurrency trading involves buying and selling digital assets like Bitcoin and Ethereum to generate profits. It can be done manually or through automated systems that execute trades based on predefined strategies.

How do machine learning models improve trading accuracy?
Machine learning models analyze historical data to identify patterns and predict future price movements. Techniques like LSTMs and Random Forests capture complex relationships, enhancing decision-making in high-volatility markets.

What are the risks associated with cryptocurrency trading?
Risks include high volatility, cybersecurity threats, regulatory changes, and scalability issues. Proper risk management strategies, such as diversification and stop-loss orders, are essential.

Can technical analysis be applied to cryptocurrencies?
Yes, technical analysis tools like moving averages, RSI, and Fibonacci retracements are commonly used to identify trends and generate trading signals in cryptocurrency markets.

What role do cryptocurrency exchanges play?
Exchanges provide platforms for buying, selling, and trading cryptocurrencies. They vary in features, supported assets, and regulatory compliance, influencing factors like liquidity and security.

How is sentiment analysis used in trading?
Sentiment analysis evaluates public opinion from social media and news sources to gauge market mood. This data, combined with machine learning, can predict price movements and inform trading decisions.

👉 Access real-time market analysis tools

Conclusion

Cryptocurrency trading represents a dynamic and rapidly evolving field. This survey highlights the diversity of research, from trading systems and strategies to risk management and market analysis. Emerging technologies like machine learning and econometrics continue to enhance predictive accuracy and strategic development. However, challenges such as volatility, security, and regulatory uncertainty remain. Future research should focus on refining models, exploring new data sources, and developing robust frameworks for sustainable trading practices.