Dates and Locations

23 & 24 Nov '10	London, UK	Apex City of London Hotel >>
20 & 21 Jan '11	Hong Kong	Sheraton Hong Kong Hotel and Towers >>

Please visit our Course Diary to see dates and locations for all of our UK and overseas courses.

Course Outline: Backtesting and Optimization Workshop

This two day workshop is designed for traders and investment managers of all asset classes and time scales who are looking for guidance on how to backtest their own trading strategy or model. Delegates will use MATLAB software (PC supplied) to solve backtesting problems using real market data.

A. Overview
1. What is backtesting and how does it differ from "simulations"?
2. Why is backtesting a necessary step for profitable automated trading?
3. Why is backtesting not a sufficient step to ensure profitability in automated trading?
4. What can we do to increase the predictive power of our backtest results?
5. Incorporating optimization and money/risk management schemes in your backtest
6. How to identify good/bad strategies even before a backtest

B. Choosing a backtest platform
1. Criteria for choosing a suitable backtest platform
2. A list of backtesting platforms
3. Discussion of pros and cons of each platform
4. Special note: backtesting high frequency strategies
5. Why do we choose MATLAB?

C. Choosing a historical database
1. Criteria for choosing a good historical database
2. A list of vendors of historical databases
3. Discussion of pros and cons of each database

D. Tutorial to MATLAB
1. Quick survey of syntax
2. Exercises: building some utilities useful for trading and plotting simple graphs
3. Using toolboxes
4. Datafeed toolbox:
a) Statistics toolbox; b) Financial, financial derivatives, fixed-income, econometrics, optimization, neural network, wavelet toolboxes; c) Freeware: Spatial-econometrics.com

E. Exercise: backtesting various types of strategies
1. Single ETF/stock/currency
2. Futures: a) How to handle rollovers and to create continuous contracts; b) How to backtest seasonal strategies.
3. Portfolio of stocks

F. Performance measurement
1. Returns and excess returns
2. Drawdown calculation
3. The importance of Sharpe ratio
4. Subtleties in Sharpe ratio calculation

G. Ways in which your backtest may not reflect reality
1. Look-ahead bias
2. Data-snooping bias and ways to avoid it:
a) Out-of-sample testing and why it is not a panacea; b) Cross validation; c) Randomization
3. Survivorship bias
4. Impact of noisy data on different types of strategies
5. Impact of historical or current short-sale constraint
6. Impact of transaction costs
7. Volume constraint
8. Regime change
9. The importance of paper trading

H. Optimization of parameters
1. Using MATLAB to optimize
2. Using Alphacet Discovery to optimize (Demo)
3. Interpolating the optimal parameters
4. Pitfalls of parameter optimization
5. What is "portfolio optimization" and whether you really need it

I. Money and risk management
1. Optimal capital allocation and leverage via "correlation analysis" and the Kelly formula for continuous finance