Classification of trading systems
Delve into trading's complex world with a guide on systems and methodologies, enabling tailored strategies through crucial classification
TRADING
Javier González-Barros, CFTe
11/6/20234 min read
What is a trading system?
A trading system is defined as a set of rules that generate entry and exit signals in the markets. For its effectiveness, it must integrate solid risk management, prudent monetary management, and an understanding of trading psychology (Elder, 1993, "Trading for a Living"). These components are essential for long-term success in trading.
Classification of trading systems
Mechanical Systems: Automate signal generation without human intervention once the rules are established. These systems, as described by Kaufman (2013) in "Trading Systems and Methods," send orders directly to brokers, facilitating efficient execution and reducing the emotional impact on decision-making.
Discretionary Systems: Combine established rules with the trader's intuition for making decisions. This flexibility allows for adaptation to market changes, though it requires greater emotional management, as explained by Douglas (2000) in "Trading in the Zone."
Based on the data used and market permanence:
Continuous Systems: Are always in the market, adjusting positions according to conditions to maximize returns at the cost of higher risks.
Pure Intraday and Continuous Intraday Systems: Vary in risk and return, with the former seeking to minimize overnight risks and the latter seizing opportunities throughout the day.
According to Charlie F. Wright's theory:
Trend Systems: Benefit from prolonged market movements.
Counter-Trend Systems: Leverage market reversals in periods of consolidation.
Volatility Expansion Systems: Seek gains in volatile markets.
1. Lunar Phase Analysis
Foundation: This approach is based on the theory that lunar phases can influence human behavior, which in turn can affect market movements. The premise is that the psychological impact of the moon's phases can lead to systematic patterns in financial markets.
Advantages: Unique perspective on market prediction that goes beyond traditional financial indicators.
Challenges: Skepticism in the academic community due to its unconventional nature. Difficulty in quantifying the impact.
Sources: Dichev and Janes (2003) conducted a study titled "Lunar Cycle Effects in Stock Returns," published in the "Journal of Private Equity," which found that stock returns are lower on the days around a full moon than on the days around a new moon.
2. Econometric Models
Foundation: These models use statistical methods to forecast future market movements based on historical economic and financial data. They can include a variety of economic indicators, from GDP growth rates to interest rates, to model the market behavior.
Advantages: Provides a quantitative and theoretical foundation for predicting market movements. Can incorporate a wide range of economic indicators.
Challenges: High complexity and requires deep understanding of both economic theory and statistical methods. Models may be overly reliant on historical data, which can limit their predictive power in unprecedented market conditions.
Sources: Brooks (2014) in "Introductory Econometrics for Finance" provides comprehensive coverage on using econometric models in financial market analysis, offering insights into how these models are constructed and applied.
3. Neural Networks
Foundation: Neural networks are a subset of machine learning designed to recognize patterns by mimicking the way human brains operate. In trading, they are used to predict market movements by analyzing large datasets and identifying complex patterns not easily detectable by humans.
Advantages: Ability to process and learn from vast amounts of data. Can adapt to new information, improving their predictions over time.
Challenges: Requires extensive data for training. The "black box" nature of neural networks can make it difficult to interpret how decisions are made.
Sources: Zhang, Patuwo, and Hu (1998) in "Forecasting with Artificial Neural Networks: The State of the Art," published in the "International Journal of Forecasting," discuss the application of neural networks in predicting financial markets.
4. Machine Learning
Foundation: Machine learning involves algorithms that enable computers to learn from and make decisions based on data. In trading, machine learning algorithms analyze market data to find investment opportunities and predict market trends.
Advantages: Flexibility to learn from data and improve predictions over time. Can analyze a broader range of data types than traditional statistical models.
Challenges: Complexity in development and requirement for continuous data input for learning. Risk of overfitting to historical data.
References
Brooks, C. (2014). Introductory Econometrics for Finance. Cambridge: Cambridge University Press.
Dichev, I. D., & Janes, T. D. (2003). Lunar Cycle Effects in Stock Returns. The Journal of Private Equity, 6(4), 8-29.
Douglas, M. (2000). Trading in the Zone: Master the Market with Confidence, Discipline, and a Winning Attitude. New York: Prentice Hall.
Elder, A. (1993). Trading for a Living: Psychology, Trading Tactics, Money Management. New York: John Wiley & Sons.
Kaufman, P. J. (2013). Trading Systems and Methods. New Jersey: John Wiley & Sons.
Lefèvre, E. (1923). Reminiscences of a Stock Operator. New York: George H. Doran Company.
Murphy, J. J. (1999). Technical Analysis of the Financial Markets: A Comprehensive Guide to Trading Methods and Applications. New York: New York Institute of Finance.
Prado, M. L. (2018). Advances in Financial Machine Learning. New Jersey: John Wiley & Sons.
Tharp, V. K. (2010). Trade Your Way to Financial Freedom. New York: McGraw-Hill.
Zhang, G., Patuwo, B. E., & Hu, M. Y. (1998). Forecasting with Artificial Neural Networks: The State of the Art. International Journal of Forecasting, 14(1), 35-62.
