Key Takeaways
- Algorithmic trading employs process- and rules-based computational formulas for executing trades.
- Black-box or profit-seeking algorithms can have opaque decision-making processes, raising concerns among policymakers and regulators.
- The practice has grown significantly since the early 1980s, primarily used by institutional investors and large trading firms.
- Although it provides advantages like faster execution and reduced costs, it can also exacerbate market issues, causing flash crashes and immediate liquidity loss.
Algorithmic trading uses advanced mathematical models along with some human oversight to make trade decisions. These decisions can be split-second thanks to high-frequency trading (HFT), allowing firms to execute tens of thousands of trades per second. This technology can be used for order execution, arbitrage, and trend trading strategies.
Evolution of Algorithmic Trading
The use of algorithms in trading rose after computerized trading systems were introduced in American financial markets during the 1970s. By 2009, over 60% of trades in the U.S. were executed by computers.
HFT was brought into the public spotlight by Michael Lewis’ bestseller Flash Boys, which revealed that companies were in an arms race to develop ever faster computers to gain split-second advantages over rivals.
The Diverse Types of Algorithmic Trading
Various algorithmic strategies are designed to make trading decisions automatically. Here are some prominent types:
-
Arrival Price Algorithms: Execute trades close to the stock price when the order was placed, minimizing market impact.
-
Basket Algorithms: Also known as portfolio algorithms, these take into account the effects on other decisions and securities in a portfolio.
-
Implementation Shortfall Algorithms: Aim to minimize the cost of executing an order when it differs from the decision price.
-
Percentage of Volume: Adjusts order sizes based on real-time market trading volume to maintain a preset market volume percentage.
-
Single-Stock Algorithms: Optimize the trade execution of a single security, factoring in market conditions and order size.
-
Volume-Weighted Average Price (VWAP): Executes orders at a price that closely matches the stock’s volume-weighted average price over a specified period.
-
Time-Weighted Average Price (TWAP): Distributes trades evenly across a set period to minimize market disruption.
-
Risk-Aversion Parameter: Adjust trading aggressiveness based on the trader’s or client’s risk tolerance.
Example of Algorithmic Trading
Consider an algorithm designed to buy 100 shares of Company XYZ whenever the 75-day moving average surpasses the 200-day moving average, a bullish crossover indicating an upward trend. The algorithm monitors these averages, executing the trade automatically when the condition is met, ensuring precise, emotion-free trading.
Black Box Algorithms
Unlike other algorithms that follow predefined rules, black box algorithms autonomously determine the best path to achieve set objectives, based on market conditions and external events. They utilize AI and machine learning, making their decision-making processes obscure even to their designers.
This opaqueness raises questions about accountability and risk management. Despite these concerns, black box algorithms remain popular in high-frequency and advanced investment strategies due to their adaptive capabilities and performance.
Embracing Open Source in Algorithmic Trading
Open-source platforms have democratized algorithmic trading, allowing individual traders and amateur programmers to bring new ideas. Some firms host competitions for programmer-generated algorithms, rewarding the most profitable ones.
The Fintech Open Source Foundation reported that a growing number of financial professionals are contributing to and using open-source data science and AI/ML platforms, despite concerns over safeguarding proprietary knowledge.
Pros and Cons of Algorithmic Trading
Advantages
-
Speed & Efficiency: Algorithms execute trades faster and more efficiently than human traders.
-
Accuracy: Reduced chances of manual errors.
-
Emotionless Trading: Eliminates emotional and psychological factors from the decision-making process.
-
Backtesting: Strategies can be tested against historical data before real-world application.
-
Anonymity & Market Access: Trades are automated, providing quicker and anonymous access to markets.
Disadvantages
-
System Reliability: Over-reliance on technology can lead to significant losses if systems fail.
-
Complexity: The myriad of algorithms and jargon can be challenging.
-
Regulatory Challenges: Constantly evolving regulation requires continuous monitoring.
-
Market Volatility: Algorithms can exacerbate market volatility and cause liquidity issues.
Getting Started with Algorithmic Trading
To begin, learn programming languages like C++, Java, or Python, and gain a solid understanding of financial markets. Create or choose a trading strategy and backtest it using historical data. Most brokerage firms offer platforms for this purpose, and open-source communities provide a wealth of shared software and advice for novices.
The Cost of Algorithmic Trading
The required capital can vary significantly based on the chosen strategy, brokerage, and market.
High-Frequency Trading vs. Algorithmic Trading
HFT is a subset of algorithmic trading characterized by extremely high speed and a large number of transactions, leveraging high-speed networking and computing capabilities.
Summary
Algorithmic trading offers multiple benefits, including speed, efficiency, and objectivity in trading decisions, but it also carries risks such as systemic risk and technical glitches. Adapting to an ever-evolving technological and regulatory landscape is crucial for its continued success.
Related Terms: automated trading, fintech, artificial intelligence, machine learning, backtesting, stock market.
References
- R. Kissell. Algorithmic Trading Methods: Applications Using Advanced Statistics, Optimization, and Machine Learning Techniques. Elsevier, 2020. Pages 23-39.
- Securities and Exchange Commission. “Release No. 34-59593; File No. NYSEALTR-2009-28”, Page 3.
- Deutsche Bank Research. “High-Frequency Trading: Reaching the Limits”, Page 2.
- Michael Lewis. Flash Boys. W. W. Norton, 2015.
- R. Kissell. Algorithmic Trading Methods: Applications Using Advanced Statistics, Optimization, and Machine Learning Techniques. Elsevier, 2020. Page 32-33.
- R. Kissell. Algorithmic Trading Methods: Applications Using Advanced Statistics, Optimization, and Machine Learning Techniques. Elsevier, 2020. Page 33.
- E. Tsang. AI for Finance. Taylor & Francis Group, 2023. Pages 2-11.
- Jarek Gryz and Marcin Rojszczak. “Black Box Algorithms and the Rights of Individuals: No Easy Solution to the ‘Explainability’ Problem”. Internet Policy Review. (June 30, 2021.)
- Rola Shaway, et al. Artificial Intelligence in Financial Services: Advantages and Disadvantages. In J. S. Hasan, et al., eds. Artificial Intelligence for Capital Markets. CRC Press, 2023. Pages 28-39.
- Mahmoud El Samad, et al. Machine Learning and Big Data in Financial Services. In J. S. Hasan, et al., eds. Artificial Intelligence for Capital Markets. CRC Press, 2023. Pages 13-27.
- Two Sigma. “Open Source”.
- Fintech Open Source Foundation. “2023: State of Open Source in Financial Services”. Page 30.
- Bank of England. “Judgement Day: Algorithmic Trading Around the Swiss Franc Cap Removal”, Pages 24-25.
