Chan, Quantitative Trading

Ernest Chan’s Quantitative Trading—How to Build Your Own Algorithmic Trading Business (Wiley, 2009) was not written for me so I won’t do it the disservice of reviewing it. But I’ve been intrigued with high-frequency trading strategies ever since I first heard of the phenomenal returns of James Simons’ Renaissance Technologies. My decidedly “low-frequency” programming skills are just one reason I could never be a high-frequency trader. The list goes on and on. The fascination remains.

Chan devotes only a couple of pages to high-frequency trading, but I think they’re interesting enough to share. The main appeal of high-frequency trading strategies is that they can achieve very high Sharpe ratios. That is, they can produce outsized returns per unit of risk. It all goes back to the law of large numbers. The more trades, the smaller the deviation from the mean return. As long as the strategy is sound (the mean return is positive) high-frequency trading grinds out results day in and day out that deviate only modestly from the mean. With this level of predictability the trader can boost leverage and reach for the sky.

What are the problems with high-frequency trading? The first, and most obvious, is that the trading strategy has be to rock solid, normally exploiting tiny inefficiencies in the market or providing temporary liquidity for a fee. Moreover, for any Goldman or Renaissance wannabe, transaction costs would undoubtedly be a nonstarter. In brief, for the retail trader high-frequency trading, at least in its “on steroids” form, is out of reach.

By the way, the question with which I began the last paragraph is increasingly being raised in a different context. The Zero Hedge blog recently posted a mini white paper entitled “Why Institutional Investors Should Be Concerned about High Frequency Traders" (http://zerohedge.blogspot.com/).