So I’ve been trying to push my self into writing a book review of “High-Frequency Trading: A Practical Guide to Algorithmic Strategies and Trading Systems“, written by Irene Aldridge. But I’m not really a book review kind of guy, so I’ve been putting it off. It’s been ages since I last wrote a book review, so I thought I would link to a few reviews and then give a few notes instead.
In general I view this book as an intro book. It eases the readers journey by starting out with the evolution and business aspects of HFT, before diving into the more technical parts. It also has a metric ton of references throughout the book for those interested in a deep dive.
I do have a low frequency trading system up and running, so based on that I would recommend this book to people interested in both high and low algorithmic trading as there is definitely a lot of cross interest. Chapter 5 for instance, “Evaluating Performance of High-Frequency Strategies”, covers techniques like Sharpe and Treynor ratios, Jensen’s Alpha and a buch more. These are topics of both high and low frequency interest.
Another chapter I particularly enjoyed was the Risk Management chapter, giving the reader a buch of hands on and practical examples on how this can be performed.
Chapter 15 covers back-testing of trading models, basically divided into two types of strategies: Point forecasts and directional forecasts. What I find very strange about this chapter is how directional forecasts are evaluated. In essens, in addition to a stop-loss level, a take-profit level is also put into each signal/order. Now in my world, adding a take-profit to a directional forecast would change that forecast to a point forecasts. I see the benefit of having a stop-loss, but unless that is triggered I do not believe a directional forecast should be closed till the direction (trend) changes. Of course there are operational benefits of using a take-profit (in addition to a stop-loss) level for each order, basically making them a fire and forget type of order, but still, IMO, directional forecasts and their orders should be just that, directional.
All in all I would recommend this book to anyone interested in an introduction to HFT, tips and references to further reading. As said, I would also recommend this book to anyone doing algorithmic LFT. There’s so much crap being said about HFT, so it’s good to get some pure facts on the table every once in a while.