canceled IB Collegiate Trading Olympiad ...picking up the slack

Hi,

I would like to first introduce myself. I am an employee with the Buttonwood Group Trading Firm. We do a lot of quant work, so we often visit this site (not to spy, but just to get an idea of what students are thinking about)

I noticed that the IB trading competition was well received in the past, but recently canceled. Our firm would like to implement something in its place for communities like Quantnet.

Any input is welcome :)
 
I looked pretty closely at how the funding was distributed. The grand prize was very nice, but prizes were also given for people who made $500 profit on millions of initial funds. We won't be doing that :)

However, we would like your input on collaborative tools. We want teams rather than individuals as well.

We will probably offer prize money and jobs/contracts.

more input needed!
 
Guaranteed jobs for the top 3, 5 performers, GC sponsor available
People trade individually or part of a team of X members from University Y
Open to undergrad, graduate students as well as the recently graduated
Time the trading period so that it does not conflict with school finals (IB Olympiad schedule isn't ideal as it started around Dec)
Have a longer period where people with paper account can test their algo prior to the real trading
Have real time display of top performers.
 
Have real time display of top performers.
It would be nice to study the impact of the real time leader board on trading strategies with the sole parameter for best being the overall return compounded by the fact that only a few people get the job/prizes.

I think small prizes for good performers would not only be in our selfish interest, but also help attract more participation. Furthermore, somehow we should avoid a repetition of IB2009 whereby an individual traded only 1 week to get > 200% profit. (I do mean an incorporation of Risk, Maximum Drawdown etc. for a complete ordering of the participants performance)
 
Guaranteed jobs for the top 3, 5 performers, GC sponsor available

Come on Andy, be serious. This is a ridiculous expectation for a prize. I liked the IB competition because they gave prizes to a large number of competitors as long as they had a positive return. I would be happy with a smaller prize pool than IB along with one that was more evenly distributed among the top competitors. Give a prize to anyone with a positive return and above a benchmark index. Encourage and reward participation. It takes a tremendous amount of effort to compete in an automated trading competition.

Time the trading period so that it does not conflict with school finals (IB Olympiad schedule isn't ideal as it started around Dec)

The IB comp started mid January. The deadline for submitting strategies was in December but they were accepting applications back in September. There was plenty of time for people to get their act together before finals.

Have a longer period where people with paper account can test their algo prior to the real trading

For people that got into the IB comp early, they had over a semester to test their strategies.
 
In terms of having a job based prize along with the monetary prize, what would you guys think about some sort of study abroad prize as well, where you went to cambridge to work with quant profs?

would you be more or less interested if this was hosted along with a university?
 
Just start the damn competition, so us wannabe quants can slaughter each other!:smt006

Seriously, study abroad prize, working with a prof seems very nice, but you can easily get into problems with it, lets suppose a guy, from a country you have never heard of (like me :D), wins the competition. That would mean you would need to acquire a visa, work permit or whatnot document, for that person. I see major problems.
Of course, you could discriminate the competitors and say:"Well, sorry you're not from XY country. No prizes for you.". I hope that doesn't happen.

Best stick to monetary prizes, maybe a job or an internship.
 
I see that Ramnik has already touched this point, but can I also re-emphasize the importance of incorporating VaR in your final analysis. 'Return only' rankings encourage unrealistic risk taking, especially by candidates with a weak algorithm towards the end of the trading period.

I haven't previously heard about the IB olympiad, is it also possible to open this up for prospective students for 2010 term?
 
In terms of having a job based prize along with the monetary prize, what would you guys think about some sort of study abroad prize as well, where you went to cambridge to work with quant profs?

I do not think it would be feasible for many of us to claim such prize. We have our schools, jobs, schedules, CFA certifications, etc.
 
In terms of having a job based prize along with the monetary prize, what would you guys think about some sort of study abroad prize as well, where you went to cambridge to work with quant profs?

i agree with people who mentioned logistical/legal hurdles with this option.
Additionally, there's an issue with "why?". It's not like NYC lacks top talent (professional or academic) and people need to go elsewhere to tap it (a la going to Italy to learn landscape painting). if anything, we could be the ones offering study abroad opportunities to our cambridge colleagues ;]
 
The best part of the IB competition was the fact that the “real” TWS trading platform and market data was used which gave the participants a real-world trading desk feel. By the end of the two month competition I was stressed by my additional part time job but quite happy the prize covered my educational expenses.

My comments are as follows:
--- The need for a realistic platform in a paper trading account is not trivial. IB owns the platform, market data and paper-trading infrastructure so they were able to host the contest without huge additional costs beyond the prize money.

--- The comments above on making the outcome dependent on the amount of risk taken should be considered as I felt there were more then a few “roll the die” strategies in play. As it costs zero to enter, a player that bets the whole account on option premium will be spectacularly right or spectacularly wrong, which costs nothing in this case. I had stripped most of the risk parameters from one of the futures strategies I use to this day and while I had good performance I felt the competition provided an advantage to the gamblers.

--- The prize money helps justify the effort so while the very bottom prizes were give aways I felt the $10,000 prizes were very important to keep people interested in the competition as it does require substantial effort.

--- What are Buttonwood Group’s goals for the proposed contest? If you intend to identify people who can code to an API, work with market data, and code trading strategies your rules will be different then if you are looking for people who can trade. There are many commercial packages which integrate into trading platforms that can be used by people who can trade but are not necessarily experts at coding or data manipulation.

Let us know if you decide to proceed with the contest as a trip to Baruch would probably be very helpful to the students and the alumni would be able to provide additional feedback.
 
Top