Quantitative Trading: How to Build Your Own Algorithmic Trading Business

by Ernest P. Chan

Quick Summary

A practical guide for retail traders on building algorithmic trading systems, covering strategy identification, backtesting with MATLAB and Excel, automated execution, money management, and scaling strategies based on real-world profitability.

Detailed Summary

Ernest P. Chan, a former quantitative researcher at Morgan Stanley and Credit Suisse turned independent trader and consultant, provides a practical, step-by-step guide for retail traders aspiring to build algorithmic trading businesses. The book distinguishes itself from academic treatments by focusing on immediately actionable implementation.

Chan begins by establishing that retail quantitative traders can compete with institutional players by exploiting several structural advantages: freedom from investor relations obligations, ability to trade small and illiquid instruments, flexibility in holding periods, and the absence of bureaucratic constraints on strategy implementation. He then outlines the process of identifying viable strategies, emphasizing the importance of finding ideas through academic journals, textbooks, and financial blogs rather than through pure data mining.

The backtesting methodology sections are rigorous and practical, covering implementation in both MATLAB and Excel. Chan addresses common pitfalls including look-ahead bias, survivorship bias, data-snooping bias, and the critical distinction between in-sample and out-of-sample performance. The concept of the Sharpe ratio as the primary performance metric is explained, along with its limitations and proper interpretation.

Strategy types are surveyed including mean-reversion strategies (pairs trading, statistical arbitrage), momentum/trend-following strategies, and high-frequency market-making approaches. For each category, Chan discusses the theoretical basis, typical implementation approaches, common pitfalls, and realistic performance expectations. The Kelly criterion for optimal position sizing is explained and modified for practical trading applications.

The execution section covers building automated trading systems using MATLAB and connecting to Interactive Brokers through their API. Chan discusses the tradeoffs between different programming languages and platforms. The money management chapter addresses portfolio-level risk management, including the relationship between leverage, expected returns, and drawdown probability.

The final chapters address scaling a trading operation, including when to add capital to profitable strategies, when to retire degrading strategies, and how to evaluate whether poor recent performance reflects normal variance or fundamental strategy decay.