- Deploying Computer Algorithms to Conquer the Markets -
by Ernest P. Chan
Purchase Now !
Just with Paypal
Just with Paypal
Book Details
Price
|
2.50 |
|---|---|
Pages
| 267 p |
File Size
|
1,429 KB |
File Type
|
PDF format |
ISBN
| 978-1-119-21960-6 (Hardcover) 978-1-119-21967-5 (ePDF) 978-1-119-21965-1 (ePub) |
Copyright©
| 2017 by Ernest P. Chan. |
Ernest P. Chan has been the managing member of QTS Capital Management,
LLC, a commodity pool operator and trading advisor, since 2011.
An alumnus of Morgan Stanley and Credit Suisse, he received his PhD in
physics from Cornell University and was a researcher in machine learning at
IBM’s T.J.Watson Research Center before joining the financial industry. He
is the author of Quantitative Trading: How to Build Your Own Algorithmic Trading
Business and Algorithmic Trading: Winning Strategies and Their Rationale. Find
out more about Ernie at www.epchan.com.
P R E FACE
The best way to learn something really well is to teach it to someone else
(Bargh and Schul, 1980). So I confess that one major motivation for my
writing this book, the third and the most advanced to date in a series, is to
force myself to study in more depth the following topics:
■ The latest backtesting and trading platforms and the best and most
cost-effective vendors for all manners of data (Chapter 1);
■ How to pick the best broker for algorithmic executions and what precautions
we should take (Chapter 1);
■ The simplest way to optimize allocations to different assets and strategies (Chapter 1);
■ Factor models in all their glory, including those derived from the options
market, and why they can be useful to short-term traders (Chapter 2);
■ Time series techniques: ARIMA, VAR, and state space models (with hidden
variables) as applied to practical trading (Chapter 3);
■ Artificial intelligence/machine learning techniques: particularly methods
that will reduce overfitting (Chapter 4);
■ Options and volatility trading strategies, including those that involve
portfolios of options (Chapter 5);
■ Intraday and higher frequency trading: market microstructure, order
types and routing optimization, dark pools, adverse selection, order
flow, and how to backtest intraday strategies with tick data (Chapter 6);
■ Bitcoins: bringing some of the techniques we covered to this new asset class (Chapter 7);
■ How to keep up with the latest knowledge (Chapter 8);
■ Transitioning from a proprietary trader to an investment advisor (Chapter 8).
I don’t know if these topics will excite you or bring you profits, but my
study of them has certainly improved my own money management skills.
Besides, sharing knowledge and ideas is fun and ultimately conducive to creativity and profits.
You will find most of the materials quite accessible to anyone who has
some experience in a quantitative field, be it computer science, engineering,
or physics. Notmuch prior knowledge of trading and finance is assumed
(except for the chapter on options, where we do assume basic familiarity).
However, if you are completely new to trading, you may find my more
basic treatments in Quantitative Trading (Chan, 2009) and Algorithmic Trading
(Chan, 2013) easier to understand. This book can be treated as a continuation
of my first two books, with coverage on topics that I have not discussed
before, but it can also be read independently.
Although many prototype trading strategies have been included as
examples, one should definitely not treat them as shrink-wrapped products
ready to deploy in live trading. As I have emphasized in my previous
books, nobody should trade someone else’s strategies without a thorough,
independent backtest, removing all likely sources of biases and data errors,
and adding various variations for improvement. Most, if not all, the
strategies I describe contain hidden biases in one way or another, waiting
for you to unearth and eliminate.
I use MATLAB for all of my research in trading. I find it extremely
user-friendly, with constantly improving and new features, and with an
increasing number of specialized toolboxes that I can draw on. For example,
without the Statistics and Machine Learning Toolbox, it would take much
longer to explore using AI/ML techniques for trading. (See why Google
scientist and machine learning expert Kevin Murphy prefers MATLAB to R
for AI/ML research in Murphy, 2015.) In the past, readers have complained
about the high price of a MATLAB license. But now, it costs only $150 for a
‘‘Home’’ license, with each additional toolbox costing only $45. No serious
traders should compromise their productivity because of this small cost.
I am also familiar with R, which is a close relative to MATLAB. But frankly,
it is no match for MATLAB in terms of performance and user-friendliness.
A detailed comparison of these languages can be found in Chapters 1 and 6.
If you don’t already know MATLAB, it is very easy to request a one-month
trial license from mathworks.com and use its many free online tutorials
to learn the language. One great advantage of MATLAB over R or other
open-source languages is that there is excellent customer support: If you
have a question, just email or call the staff at Mathworks. (Often, someone
with a PhD will answer your questions.)
I have taught many of these topics to both retail and institutional traders
at my biannual workshops in London, as well as online (www.epchan.com).
In order to facilitate lecturers who would like to use this as a textbook for a
special topics course on Algorithmic Trading, I have included many exercises
at the end of most chapters. Some of these exercises should be treated as
suggestions for open-ended projects; there are no ready-made answers.
Readers will also find all of the software and some data used in the
examples on epchan.com/book3. The userid and password are embedded
in Box 1.1. But unlike my previous books, some of the data involved in
the example strategies are under strict licensing restrictions and therefore
are unavailable for free download from my website. Readers are invited
to purchase or rent them from their original sources, all of which are
described in Chapter 1.
Table of Contents
Preface ix
CHAPTER1 The Basics of Algorithmic Trading 1
CHAPTER2 Factor Models 27
CHAPTER3 Time-Series Analysis 59
CHAPTER4 Artificial Intelligence Techniques 83
CHAPTER5 Options Strategies 119
CHAPTER6 Intraday Trading andMarket Microstructure 159
CHAPTER7 Bitcoins 201
CHAPTER8 Algorithmic Trading Is Good for Body and Soul 215
Bibliography 227
About the Author 235
Index 237
Bookscreen
I have benefited from tips, ideas, and help from many people in putting
the content together. An incomplete list would include:
■ Stephen Aikin, a renowned author (Aikin, 2012) and lecturer, who helped
me understand implied quotes due to calendar spreads in the futures markets (Chapter 6).
■ David Don and Joseph Signorelli of Lime Brokerage, who corrected some
of my misunderstanding of the market microstructure (Chapter 6).
■ Jonathan Shore, infinitely knowledgeable about bitcoins, who helped
compile some order book data in that market and shared that with me (Chapter 7).
■ Dr. Roger Hunter, CTO at our firm, QTS Capital Management, who
reviewed my manuscript and who never failed to find software bugs in my codes.
■ The team at Interactive Brokers (especially Joanne, Ragini, Mike, Greg,
Ian, and Ralph) whose infinite patience with my questions about all issues
related to trading are much appreciated.
I would like to thank Professor Thomas Miller of Northwestern University
for hiring me to teach the Risk Analytics course at the Master of
Science in Predictive Analytics program. In the same vein, I would also
like to thank Matthew Clements and Jim Biss at Global Markets Training
for organizing the London workshops for me over the years. Quite a few
nuggets of knowledge in this book come out of materials or discussions
from these courses and workshops.
Trading and research have been made a lot more interesting and enjoyable
because I was able to work closely with our team at QTS, who contributed
to research, ideas, and general knowledge, some of which find their way into
this book. Among them, Roger, of course, without whom there wouldn’t be
QTS, but also Yang, Marcin, Sam, and last but not least, Ray.
Of course, none of my books would come into existence without the
support ofWiley, especially my long-time editor Bill Falloon, development
editor Julie Kerr, production editor Caroline Maria, and copy editor Cheryl
Ferguson (from whom no missing ‘‘end’’ to a ‘‘for’’-loop can escape). It was
truly a great pleasure to work with them, and their enthusiasm and professionalism
are greatly appreciated.