From the September 01, 2011 issue of Futures Magazine • Subscribe!

True cost of learning: Model implementation

Analysis & execution

The GIRTH model, run on a spreadsheet, directs the student to implement a trade (long, short or exit) based on the open of the four-hour candlestick and analysis of the 10- and 20-period exponential moving averages (EMAs). Thus, the student must actively manage the model, watching the market at 1 a.m., 5 a.m., 9 a.m., 1 p.m., 5 p.m. and 9 p.m. To avoid the requirement that students monitor the markets 24 hours a day, they trade only at these times. This affects their entrance and exit of a trade. (It’s expected and accepted that the student-managed model will underperform an exact trade cross method because of the time situation.)

Because the spot forex market transacts 24 hours a day, from late Sunday afternoon to late Friday afternoon, students must be focused on the market and on team participation to manage and update the model, execute and inform others of decisions. FXCM is used to mark-to-market the book, running FXCM reports each time a position is closed (paper and digital). The fund also marks-to-market the book once a week, on Friday, to determine percentage returns.

The initial FFF team consisted of 20 traders, sophomores through graduate students. All students had to take a trading class either before entering the fund or simultaneously. The senior management of the simulation fund, consisting of students only, picked the 2010-11 academic year students to begin trading real money. Two students effectively left the fund in January 2011 to take full-time internships. Additionally, in May 2010, FFF "hired" junior traders to provide a pool to take on full-time trading positions in the fall of 2011. FFF is looking to provide continuity of knowledge and also to leverage the skill pool as students graduate. It is impractical and time consuming to teach the fund from ground zero every semester or every year.

FFF results are shown in "To err is expensive" (below) for Oct. 29 to May 5. These are actual trade results. FFF produced a simple return of 7.18% (including errors); this is the blue line. If the fund had no errors in execution, it would have made 18.59%; this is the orange line. Errors cost the fund approximately $285. Against an initial capital of $2,500, this error "loss" is approximately 11.5%.

<< Page 2 of 4 >>
comments powered by Disqus
Check out Futures Magazine - Polls on LockerDome on LockerDome