Customer Reviews

Quantitative Trading: How to Build Your Own Algorithmic Trading Business (Wiley Trading)

5 star:		(6)
4 star:		(5)
3 star:		(1)
2 star:		(6)
1 star:		(1)

 

 

This book lays out a road map for an independent trader to follow if they want to establish themselves as a home based quantitative trader. Whilst the title contains the words quantitative and algorithmic, it isn't heavy with high level mathematics or concepts. Instead key issues are covered in a straightforward easy to read manner.
What makes the book work is Dr Chan's personal insights from his career and experiences are intertwined throughout the topics. This leaves the reader with a great deal of confidence in his suggested approach. Most of the issues in the book are covered at a basic to intermediate level - certainly enough for you to be able to go and do further research and to know what the key issues are and what to look out for.
Criticisms - maybe the focus on Matlab, though he does recommend some cheaper clone packages that may run his example code with minor modifications.
This is a practical book. There's no pop trading psychology (though there is a psychology chapter), nor does it give you a system or rely on strategies that aren't generally available to the retail trader.
Being quantitative, strategies where discussed tend to be mean reversion or momentum based. Technical analysis isn't used as a driver for strategies. Ideas for potential quantitative strategies are discussed, but readers looking for up and running systems will need to look elsewhere.
Instead, Quantitative Trading shows the procedure for taking a strategy, combining it with correct test techniques, good execution tools and position management, and from that forming the basis for a genuinely successful home based quantitative operation.
Anyone who trades would benefit from this book as many of its discussed issues, techniques, tips and traps are universal to any trader.

Quantitative Trading: How to Build Your Own Algorithmic Trading Business (Wiley Trading), is a book written by a practitioner for practitioners. A mix of practical advices, some sample code, and a fair amount of experience, this book is a good summary, although a bit unstructured, what one needs to think about when it comes to starting your own quant trading firm or simply running your own capital in a systematic way.

While there is no ready-made, plug-and-play strategies inside the book, it does contain a number of potential strategy ideas that is worth testing. Chan is mostly focused on long/short equity strategies, mostly because that's is where he has his background. A good read, would read it again.

... but too shallow. Most of the content is just common sense, and most of the technical part, including Matlab and Excel code, of little use. Nevertheless, it gives you a blueprint of important things to take into account when going into Algorithmic Trading.

I have mixed feelings about this book, but overall, it was underwhelming. First, it really should be emphasized that this is targeted at beginners. Anyone with even a small amount of portfolio management experience will probably be familiar with the techniques discussed. Second, there is a strong emphasis on leverage and Kelly sizing (although he does suggest scaling back, eg half Kelly). Neither of these two statements are negative by themselves, but the combination seems a little dangerous. The type of person who would benefit from the information he presents probably shouldn't be taking large leveraged bets. Admittedly there are plenty of exceptions to that statement, but I think it holds in general. A good first step? Perhaps. A How-to guide for building a trading business? No way.

There doesn't seem to have been much original research conducted for the purposes of the book. He basically talks about the programs, brokerages, etc. he has used in the past and gives his opinions on a few. This is valuable to some extent, but in a book targeted at starting up a new business, I would have expected him to survey the landscape a little more. Just as an example, his code is in Matlab which he admits is probably outside the price range for many startups. Why not show the code in R? Or one of the cheaper/free Matlab clones he mentions?

There were a few technical areas I thought he breezed over too nonchalantly (assuming strategy return independence in the Kelly formula for example) that could be dangerous, but as the book is targeted at beginners, I won't hit him for those.

As a final point, the book is completely overpriced. Large font + small book + not much more than 150 pages (many of which are either code or modified entries from his blog) = where's the beef? Again, that's OK, but not in the context of a (ridiculous) $60 sticker price. Even the ~$40 at Amazon is about 2x what I believe is justified.

I was drawn to E.P. Chan's "Quantitative Trading" (2009) by a process of elimination. After losing half of my buy-and-hold retirement portfolio in the 2007-2009 bear market, I tried and rejected a variety of both fundamental and technical trading strategies. Fear and greed invariably blocked my path to success in any of these trading endeavors. Then I discovered momentum trading strategies that could be automated. Trend following strategies, such as, those proposed by Tom Lydon in "The ETF Trend Following Playbook: Profiting from Trends in Bull or Bear Markets with Exchange Traded Funds" (2009), were especially appealing to me. For example, buy when an equity's price moves above its 200 day moving average. Sell when the price falls below its 200 day moving average. What could be simpler than that? Even more appealing, however, were the relative strength, risk adjusted trading strategies that I discovered at the ETF Replay website. There I found a momentum based, quantitative, statistical model that was mechanical in operation and that separated my trading activities from my emotions, keeping fear and greed in check.

Was Chan's book written for traders like me? Not exactly. The author had in mind a reader who wants to know (1) how to start a quantitative trading business or (2) how to work as a quantitative trader at a major institution. I, on the other hand, was simply looking for ways to enhance my skills as an independent trader for managing my personal accounts.

In reading this book, I felt like a minor league player asking for help from a major league coach. Chan is a true quant with both institutional and independent trading experience. Chan offers way more expertise than I can use. For instance, Chan's book applies to trading that can be characterized as algorithmic, mean-reverting, fully automated, intraday, high frequency, leveraged, risk adjusted, and benchmarked with high Sharpe ratios and low drawdowns. My trading, on the other hand, is mechanical, not algorithmic; momentum based, not mean-reverting; semi-automated, not fully automated, monthly traded, not intraday; low frequency, not high frequency; unleveraged, not leveraged; risk adjusted and benchmarked with high Sharpe ratios and low drawdowns -- these last items being significant points of agreement with Chan.

Chan moves his readers step-by-step from determining their aptitude for quantitative trading in Chapter 1 to growing a quantitative trading business in Chapter 8. Along the way, Chan tells his readers how to select a trading strategy, how to backtest their strategy using MATLAB (or Excel), how to build an automated trading system, how to manage their money and risks, and how to refine and improve their trading strategies.

Quantitative trading is known by several other names: "algorithmic trading," "automated trading," "computer trading," and Chan's favorite, "statistical arbitrage trading." Incidentally, for statistical arbitrage trading to work, both Random Walking and the Efficient Market Hypothesis must fail.

Are there prerequisites? To benefit from Chan's book, the reader needs to have at least a first year college proficiency in statistics, algebra, and computer programming. Given this minimal background, the reader can then proceed to become an independent trader who will be able to outperform institutional money managers at their own game, namely, statistical arbitrage trading.

By the time you finish Chan's book, your statistical arbitrage trading kit will include such tools as geometric mean, moving average, standard deviation, linear regression, Gaussian distribution, mean-reverting time series, half-life time series, principal components analysis, Kelly Formula, and Sharpe ratio -- and the means to achieve a consistent monthly stream of revenue.

This is the first trading book which has pleasantly surprised me. The author does not waste time speaking to "trading psychology" or other vague concepts. He seeks to outline the issues a serious, individual trader faces -- from gathering data, to selecting trading strategies, to backtesting, to choosing a good trading platform. Plenty (prob. a bit too much) Matlab code is offered, to illustrate how to translate strategies into actual backtests. The advantage this text has, in my opinion, is that everything the author mentions is supported by specific recommendations for pursuing further. E.g., he lists specific web sites for market data and trading ideas. My only real complaint is that I would like to see a few of the topics expanded.

Very good primer on trading and quant trading. I am a beginner and found it very practical and current.
The book does not provide specific strategies (why would anyone give you their profitable strategies anyway?), but it gives guidance on how to go about creating your own strategies and trading business.
It frequently focuses on Matlab, but I just ignored the technicalities of those sections since I do not currently intend to take that route. It's quite useful, however, to get an idea of how sophisticated your competition is!

Other reviewers complained that the price is steep - this may be the case, but the book is interesting to read, and if you get only one single idea from it, it will surely be worth more that the book's cost.

I think this is a great book and being an engineer with 10+ years of Matlab I thought the code was very helpful. I can't imagine it being very helpful if you don't know Matlab or another matrix language. The book is a good introduction to quantitative trading for people unfamiliar with the subject. The author is also very active and keeps a blog and provides new material to people have purchased the book. The book has a lot of great refrences that I have turned to get more in-depth knowledge as well. I would buy this book again.

Ernest Chan's guide is inspirational and full of lesson's he's had to learn the hard way.

He covers many topics necessary to set up your own automated-trading business from home.

Although many skills are necessary for this, he covers the basics of them all without dwelling too much on excessive detail.

Like most trading books, the content is US-centric but not annoyingly so.

A very inspiring read, I now a lot more about how to measure the performance of my existing portfolio and more carefully choose how much to invest in each strategy. I also feel confident to begin my own algorithmic research and hopefully trading (currently I'm not using algorithmic methods).

I follow the author's blog although I am generally an options trader. I got the book recently to see what Chan would share beyond the blog content. As other reviewers highlighted, this is a light read to get you started. However the title is pretty self descriptive, and I didn't feel there was any misrepresentation. In this prospect, the author certainly meets expectations. The book has value overall, for stating the facts, providing codelines and punctuating the text with intelligent comments. I do appreciate that the author shares some of his personal quantitative work. The length and the writing style are appropriate for the target audience.