Backtesting and Automated Execution

• Published in: Algorithmic Trading Vol. 625; pp. 1 - 38
• Main Author: Chan, Ernie
• Format: Book Chapter
• Language: English
• Published: United States John Wiley & Sons, Incorporated 2013; Wiley; John Wiley & Sons, Inc
• Subjects: application programming interface (API); automated execution; backtesting; calendar spread; colocation; complex event processing (CEP); data‐snooping bias; FINANCE & ACCOUNTING; futures back adjustment; hypothesis testing; integrated development environment (IDE); look‐ahead bias; out‐of‐sample testing; short sales; survivorship bias; walk‐forward testing
• ISBN: 1118460146; 9781118460146
• DOI: 10.1002/9781118676998.ch1

Backtesting is the process of feeding historical data to a trading strategy to see how it would have performed. Once every detail of a strategy has been implemented as a backtest program, we can then look for pitfalls in the backtesting process. A long list of pitfalls will be examined. In any backtest, we also face the problem of finite sample size: whatever statistical measures we compute are subject to randomness. We will address this issue using a general methodology called “hypothesis testing.” Even if a backtest is done correctly without pitfalls and with high statistical significance, it doesn't necessarily mean that it is predictive of future returns. Regime shifts can spoil everything, and a few important historical examples will be highlighted. The choice of a software platform for backtesting is also an important consideration, and criteria for this choice will be discussed.