• C++ Programming for Financial Engineering
    Highly recommended by thousands of MFE students. Covers essential C++ topics with applications to financial engineering. Learn more Join!
    Python for Finance with Intro to Data Science
    Gain practical understanding of Python to read, understand, and write professional Python code for your first day on the job. Learn more Join!
    An Intuition-Based Options Primer for FE
    Ideal for entry level positions interviews and graduate studies, specializing in options trading arbitrage and options valuation models. Learn more Join!

Interview with Ernie Chan

Dr. Ernest Chan is a quantitative trader, consultant, and co-founder of EXP Capital Management, LLC., a Chicago-based investment firm. Ernie has worked for various investment banks (Morgan Stanley, Credit Suisse, Maple) and hedge funds (Mapleridge, Millennium Partners, MANE) since 1997. He received his Ph.D. in physics from Cornell University and was a member of IBM's Human Language Technologies group before joining the financial industry. He is also the author of "Quantitative Trading: How to Build Your Own Algorithmic Trading Business".

Mr. Chan - Thanks for taking the time to talk to Quant Network. What are you up to at the moment?
Thank you for having me here. I am currently dividing my time between consulting and managing my fund and other accounts.

What do you consider as your accomplishments up to this point?
Proved to myself that markets are not efficient!

What are your favorite books? Movies? TV shows? Music?
Non-fiction: everything by Malcolm Gladwell and William Poundstone. Fiction: Wolf Totem, everything by Dan Brown, and yes, Harry Potter. Movie: Godfather I, II, III. TV: I watch whatever is on, mostly PBS or Bravo. Music: Classical and some jazz.

Name the five websites do you visit regularly?
nytimes.com, newyorker.com, economist.com, theatlantic.com, globeandmail.com

What other projects are you involved in?
Giving workshops on quantitative trading in London and Hong Kong, and of course, writing my blog (epchan.blogspot.com).

Can you describe a typical work day for us?
I start the day at 7:30am, downloading data and starting up various applications to prepare for sending the day’s orders, and sometimes conferring (via phone or instant messages) with my partner in Chicago on unusual situations. In theory, all these steps can be automated, but it would require heavy investments in some software that we are not prepared to make just yet. And of course, even if they are all automated, you still want to monitor them to make sure no unexpected errors have occurred. I also try to check and reply to a few urgent emails during this time. Oh, and a 5-minute breakfast too. So this takes me to 9:30am, when the market opens, and our applications spring into action. I will sit there for maybe half an hour more to make sure the applications run normally, and then get back to doing research, strategizing with my partner over the phone, or speaking to my clients. At noon, I typically take an hour’s break to exercise, either swimming, yoga, or just hiking around the wildlife preserve near my house. I will then have my 5-minute lunch, and get back to my research and client calls. At 4pm, market closes and I shut down all the applications, and that ends the first part of my working day. I may work for another hour or so around 8pm, mostly doing research and answering emails. I also try to put in a few hours on the weekend doing the same, maybe updating my blog too.

Can you tell us the setup of your trading gig? What hardware and software do you use? Any pics to show us the gig?
The hardware in my office is really not that impressive – just 2 run-of-the-mill desktops and 2 laptops, 4 flat-panels, and a TV tuned to CNBC. Our office in Chicago is similarly equipped, and we also have a server collocated at our broker’s facilities in Jersey City. Plus we use Amazon’s computing cloud to further extend our computational resources. As readers of my book know, I use MATLAB to do practically everything, from research, downloading data, to automatically executing orders.

Going independent as a quant trader is not for everyone. There are certain technical and non-technical barriers involved. To whom would you most recommend taking this route?
Those who don’t have to count on any trading income for at least the first year of operation. Those who have at least $50,000 of cash that can be lost without significant lifestyle impact. Those who have self-discipline, emotional control, patience, and the ability to endure short-term pain. Finally, those who have basic knowledge of statistics and good programming skills, whether using Excel, VB, MATLAB, or Java.

What do you miss most about working for banks? How does being an independent trader affect your personality, health and other important parts of your life?
Flying business class on company’s expenses, free lunches (literally!) and dinners, dropping tens of thousands on software and high quality data, getting help from highly dedicated and skilled programmers and other IT professionals.
Going independent, however, make me a much happier and healthier person. I must confess that I am not a “people’s person” – I dislike being disrupted by casual conversations during work, and I particularly dislike the stress of having to please everybody in the workplace (colleagues, bosses, etc.) Here, I have practically full control over what projects I want to work on, who I want to collaborate with, and what clients I like to take on, and even who I want to talk to on a particular day. As a result, my trading performance has drastically improved. Amazingly enough, risking my own money is not very stressful once I got used to it. But I can never get used to working on projects based on the boss’ whim of the day.

Books about trading idea are dime a dozen these days. Many people have tried to run the same strategies but few succeeded. What differentiates a good trader from the bad ones?
Good traders revise and improve the strategies they read from books based on their own trading experience. Bad traders just follow those strategies blindly and give up when they are not profitable.

Cloud computing is taking off. What is your experience with the recent offerings like Amazon EC2?
I use it every day. It is a great and very economical resource.

Would the benefits of such low cost solutions be enough to entice people to move their operation online? Have you heard of any movement in that direction? Or we are still years away?
Despite my good experience, I don’t think that Amazon EC2 is suitable for high(er) frequency trading. For this type of trading, you would want to collocate your servers with your brokers or near the exchanges. The data connection speed at Amazon is no faster than an ordinary T-1 line, which is insufficient for high-frequency trading.

Say, I just finished reading your book and extremely eager to jump on the bandwagon. How and where do I start?
Find ways to improve or simplify the example strategies in my book, backtest them on your own, and open a brokerage account that allows you to submit orders via an API. And read my blog for suggestions of other resources!

What are your takes on the whole high frequency trading (HFT) craze that is being portrayed on newspaper?
For those strategies that rely on some form of front-running, then I think the bonanza will end once everybody is on the bandwagon. For those strategies that are really high-speed market-making, then I believe they will endure, though the profitability may decrease somewhat going forward.

Not so long ago, trading means equity and other vanilla financial products. Now, we have all kind of exotic derivatives that nobody understands. Technology plays an increasingly critical in the trading game. How should one prepare for the future of trading?
Keep exchanging ideas with other people and constantly explore and research new opportunities.

What is a benchmark of success for an individual trader? Taking into account the enormous personal investment put into this as opposed to being a trader at a bank, how does one justify going solo?
Isn’t happiness, serenity, and health good enough enticement?

What are the myths and misconception about quantitative trading?
The biggest myth is that once you have built a system, you can just lay back, relax, and watch millions of dollars of profits roll into your trading account. No, systems have bugs and also they become outdated very rapidly. Strategies get old and die. Furthermore, we should not risk too much equity, and therefore the profits will not be too large, but hopefully it will be steady.

There are people out there arguing that the whole finance sector contributes nothing to the society. Making money is pretty exciting and addictive but at the end of the days, what do you think you contribute to the society?
Do you believe that Harvard University contributes anything to society? What about Yale? Well, from what I heard, Harvard and Yale endowments, together with countless other universities, all invest in hedge funds. Modern economy is a gigantic web of activities. If an activity is inherently useless, I believe it will wither away to an insignificant size on its own accord.
For me personally, I contribute to the financial security of many of our investors, even if they are not necessarily the Harvard or Yale endownments.

There are so much available tools out there to do the same job. We have tools based on Matlab, C#, C++, R, Java, etc. What would you recommend for people learning the rope.
As I said, I am a MATLAB fan. R is, of course, just as good.

On a long enough timeline, the survival rate for everyone drops to zero -Zerohedge. And on your blog, you mentioned that you only became consistently profitable after 4 years of actual trading. I take it to mean that most everyone else will lose money if they can even make it past the first year. How did you manage to survive?
I believe the Zerohedge study covers only CTA’s, and actually I became consistently profitable after 1 year of trading on my own. (I doubt I will ever become profitable trading in a bank or other people’s hedge fund.) Survival depends on furiously learning from past mistakes, and in my case, the biggest past mistake is over-complexity. [/B]

What do you know now that you wish you'd known 10 years ago?
Everything should be as simple as possible, but not simpler.

Tell us something about yourself that we don’t already know.
I generally try to get by with as little work as possible.

What would you like doing 10 years from now?
Running a bigger hedge fund, and maybe a software company.