“I love it when a plan comes together.”
|Colonel John “Hannibal”
Smith of The A-Team
a/k/a George Peppard
Nice quote, but my experience is more like, “The best laid plans of mice and men oft go awry,” or possibly Murphy’s Law, “Anything that can go wrong will, and at the worst possible moment.” In most cases, things do not go as planned. Every once in a while though, things fall into place and it’s great.Take for instance some students in Tulane University’s Master of Management in Energy program. Bryce Robertson, Shanlong Zhou and Mingyu Liu were required to team up, analyze and research an energy industry issue for their Energy Projects class. This is when my phone rang. Bryce contacted me and said that they wanted to explore automated trading in the energy markets and that they want to use TT’s ADL®, also known as Algo Design Lab, to do that.
To say the least, this was music to my ears. We partner with universities so that students can experience the markets in the same way our customers do, so I’m overjoyed to see them take advantage of everything we have to offer.
The team wanted to trade December crude oil futures (CLZ3). They decided that instead of having a directional bias, they would watch the momentum of the market and trade based upon it. They chose the Relative Strength Index (RSI) to gauge momentum. The RSI is an oscillator that measures the strength of the market with readings between 0 and 100. Lower readings indicate that a down move is losing momentum (i.e., oversold), while higher readings indicate overbought conditions. They researched the crude oil futures market and determined that they would buy when the RSI dropped to 35 and sell when it hit 70. Since their strategy was short term, they wanted to keep it as simple as possible to avoid any latency issues.
They decided they would buy or sell 10 futures contracts at the market when conditions were met. An open position would be liquidated when a 10-tick profit was realized. The algorithm would monitor any positions and liquidate them when it realized a 10-tick profit. In the event this condition is never met, the position would be closed out at the end of the one-hour trading session.
ADL was then used to create the automated algorithm to execute their strategy. The tasks that needed to be completed were:
Image 1 shows how the coded their strategy using ADL:
Image 2 explains the algorithm:
In Image 2 above, the sections break out as follows:
Section 1: A live RSI feed from a one-minute bar chart of crude oil futures (CLZ3) is compared with the benchmark (35). If the live feed is less than the benchmark, it activates a market order to buy 10 CLZ3 at the market (or ask) price.
Section 2: This is the market order to buy 10 that is activated when the RSI condition in Section 1 is met. Once the order is executed (or filled), the price is sent to Section 3.
Section 3: The fill price from Section 2 is increased by 10 ticks. If it is equal to current market bid price, it will activate a market order in Section 4.
Section 4: The market order to cover the position.
They traded three one-hour sessions and made a profit of $3,460. The sample size here is somewhat small and could use further testing. Even the students noted, “Although profitable, we know that these results can be misleading due to the fact that we have not tested the algorithm in diverse markets, and we have only tested the algorithm for a limited amount of time.”
Given that this is the first time these students have explored automated trading and the first time they’ve used ADL, their efforts are noteworthy.
Is there more they could have done with ADL? Sure, but look at what they did do:
When all is said and done, it’s the topic, the tools and the teachers that excite, empower and educate the students. Curriculum that is relevant and interesting coupled with the tools to explore give students what they need to understand and experience. When all the pieces fall into place, the result is an engaged student, relevant curriculum and graduates in demand.
Posted by: Leo Murphy, TT CampusConnect™ Program Manager