Trading Systems 2nd edition

Disclaimer
Acknowledgements

Preface
Preface to the Second Edition
What has happened since 2008?
How this book is presented
The LUXOR system – when performance decays
The importance of controlling losses
Examples of working trading systems
Software and data

Part I: A Short Practical Guide to Trading System Development and Evaluation
1: What is a Trading System?
1.1 An easy example of a trading system
1.2 Why you need a trading system
1.3 The science of trading systems

2: Design, Test, Optimisation and Evaluation of a Trading System
2.1 Design
Getting started
The programming task
Which timeframe to trade?
2.2 Test
The importance of the market data
The length of your back-testing period
Rule complexity and degrees of freedom
2.3 The forecasting power of a trading system
Optimisation
Walk-forward analysis
Robustness
2.4 Evaluation of a trading system
What to look for in an indicator
Average trade
Percentage of profitable trades
Profit factor
Drawdown
Time averages

2.5 Conclusion

Part II: Trading System Development and Evaluation of a Real Case
3: How to Develop a Trading System Step-By-Step – Using the Example of the British Pound/US Dollar Pair
Introduction
3.1 The birth of a trading system
The free LUXOR system code
The entry logic
3.2 First evaluation of the trading system
Calculation without slippage and commissions
Calculation after adding slippage and commissions
3.3 Variation of the input parameters: optimisation and stability diagrams
What does stability of a system's input parameter mean? A short theoretical excursion
Dependency of main system figures on the two moving averages
Result with optimised input values
3.4 Inserting an intraday time filter50 Finding the best entry time
Result with added time filter
3.5 Determination of appropriate exits – risk management
The concept of Maximum Adverse Excursion (MAE)
Inserting a risk stop loss
Adding a trailing stop
Looking for profit targets: Maximum Favourable Excursion (MFE)
Summary: Result of the entry logic with the three added exits
How exits are affected by money management
3.6 Summary: Step-by-step development of a trading system

4: Two Methods for Evaluating the System's Predictive Power
4.1 Timescale analysis
Changing the compression of the price data
LUXOR tested on different bar compressions
Net profit and maximum drawdown dependent on the traded bar length
Explanation for the time dependency of the system
4.2 Monte Carlo analysis
The principle of Monte Carlo analysis
Exchanging the order of the performed trades
Probabilities and confidence levels
Performing a Monte Carlo analysis with the LUXOR trading system
Limitations of the Monte Carlo method

5: The Factors Around Your System
5.1 The market's long/short bias
The trend is your friend?
Consequences for system development
5.2 Out-of-sample deterioration
A Bollinger Band system with logic and code
Entry logic: Bollinger Band system
Optimising the Bollinger Band system
Out-of-sample result
Reasons for the out-of-sample deterioration
5.3 The market data bias
Expanding the training period
Conclusion: How to choose your training data
5.4 Optimisation and over-fitting
Step-by-step optimisation of the LUXOR system
Results depending on the number of optimised parameters
The meaning of the trading system's complexity
5.5 Rule complexity explained with polynomial curve fitting
Interpolating data points with polynomial functions
Predictive power of the different polynomials
Conclusions for trading system development
5.6 Example of a simple, robust trading system
Idea
Trading rules
Optimisation of the main parameter of the system
Results
Strategy evaluation

6: Periodic Reoptimisation and Walk-forward Analysis
6.1 Short repetition: normal, static optimisation
6.2 Anchored vs. rolling walk-forward analysis (WFA)
6.3 Rolling WFA on the LUXOR system
Periodic optimisation of the two main system parameters
Out-of-sample test result
Conclusion
6.4 The meaning of sample size and market structure

7: Position Sizing Example, Using the LUXOR System
7.1 Definitions: money management vs. risk management
Risk management (RM)
Money management (MM)
7.2 Application of different MM schemes
Reference: The system traded with one lot
Maximum drawdown MM
Fixed fractional MM
Fixed ratio MM
7.3 Monte Carlo analysis of the position sized system
7.4 Conclusion

Part III: Systematic Portfolio Trading
8: Dynamic Portfolio Construction
8.1 Introduction to portfolio construction
A list of the main available software for portfolio trading
The role of correlations
Publications and theoretical tools
Portfolio trading in practice
Total vs. partial equity contribution
8.2 Correlation among equity lines
8.3 A dynamic approach: equity line crossover
8.4 How to transform an average system portfolio into a profitable one: the case of LUXOR
8.5 Dynamic portfolio composition: the walk-forward analysis activator
8.6 Largest losing trade/largest losing streak/largest drawdown

9: Trading a Portfolio of Stocks
9.1 Modifications when applying a Bollinger Band system to stocks
The Bollinger Band system – trading logic for stocks
9.2 Examples: Results on single stocks
9.3 Survivorship bias in portfolio back-tests
9.4 Application of the strategy to a portfolio of S&P 500 stocks
9.4.1 The S&P 500 index
9.4.2 Testing a portfolio of S&P 500 stocks
9.4.3 Introducing a market filter
9.4.4 Optimising two parameters together: index filter and Bollinger Band length
9.4.5 Variation of the entry and exit point = upper and lower band distance
9.4.6 Ranking methods in case multiple signals occur on the same day
9.4.7 Conclusion: Development of a trading system on the S&P 500 stocks
9.5 Monte Carlo analysis of the Bollinger Band system
9.6 Periodic reoptimisation/walk-forward analysis (WFA)
Conclusion
9.7 Results on other portfolios of stocks
9.7.1 Nasdaq 100 – technology at work
9.7.2 The S&P 1500 – a large index\
9.7.3 Overview: Bollinger Band System vs. buy and hold
9.8 Position management with five stocks on the Nasdaq 100 – practical example
9.9 The psychological phenomenon of loss aversion
9.10 Different ways of daily execution
a) Professional back-testing and daily implementation, all in one
b) Professional back-testing and daily implementation, separated
c) Daily implementation without back-testing
Conclusion
Rule complexity
Testing
Optimisation
Monte Carlo analysis
Portfolio building
Dynamic risk management
Money management

Appendices: Systems and Ideas
Appendix 1. Bollinger Band System
1.1 Idea
1.2 Entry logic and EasyLanguage code
1.3 Application of the strategy to seven markets with same parameters
1.4 Results and conclusions

Appendix 2. The Triangle System
2.1 Idea
2.2 Programming and coding
2.3 Application to different liquid futures markets with same parameters
2.4 Advantages in building a portfolio
2.5 Conclusion

Appendix 3. Portfolios with the LUXOR Trading System
3.1 Idea
3.2 The trading logic
3.3 Results in the bond markets
3.4 Diversification with other market groups
3.5 Conclusion

Appendix 4. EasyLanguage Code Section
4.1 Beginning of Month (chapter 5.6)
4.2 Bollinger Band System (chapter 9)

Appendix 5: AmiBroker Code Section
5.1 Beginning of Month
5.2 Bollinger Band System (chapter 9)

Bibliography
Index