As part of a team from the Griff Investment Fund, I was lucky enough to be selected to travel to Atlanta, Georgia, to present a hedge fund trading strategy at the J Mack Robinson College of Business at Georgia State University, where our team finished second overall, out of the five finalists. 60 universities had entered, including both Harvard and MIT, and securing second meant that we won a collective $5,000 prize.
The University of York was the only UK university to enter the competition and probably the first UK university ever to do so. We were the first non-American university to get to the final, and therefore the first ever non-American university to place in the top three and secure a prize. Our team were the only undergraduate students at the final, with the other teams all being comprised of masters students.
Our idea was to adapt a well-known strategy, called the Piotroski F-Score. This considers companies based on nine indicators, and gives a binary score of 1 or 0 for each, leading to a total score of up to 9 based on the perceptions of how well a company is doing. Applying this to the US Index, the S&P 500, which is comprised of roughly 500 companies, and buying winners with a score of higher than six on a yearly basis, Piotroski’s strategy managed to return 4217% on the initial investment over 75 quarters.
Our underlying hypothesis was that the Piotroski F-Score was too focussed on companies that were already profitable, so it failed to take advantage of companies that ended up in a much better position at the end of the year from where they started. Our G-Score strategy aimed to put a higher level of emphasis on operational improvements, as empirically, this seems to yield higher share price returns.
Our new model looked to evaluate on a quarterly basis, as companies produce financial updates more frequently than on a yearly basis, so this can be taken advantage of more quickly in our modelling. We also looked to place a weighting on our indicators, that would give a higher reward to companies for satisfying for those that are perceived to be more significant. To run a model, we downloaded available public data form Thomson Reuters Datastream, and used Microsoft Excel to calculate changes, returns, and scores, which would allow us to see how our backtested data would do over the years in question.
Our new total score was out of 11.6, and we found that our optimum score threshold was now 9.5. This threshold strikes a balance between number of stocks held, and high returns, and by using it we managed to significantly beat the Piotroski F-Score, gaining 33,200% with our G-Score strategy across the same 75 quarters. We took on slightly more risk, but equally ran other models using our data, to show that with comparative risk, we still convincingly beat Piotroski. To hedge against risk when the market takes a downturn, we also looked to introduce shorts in the market, where we would sell our identified losers betting on them falling in price in the future, which would lower our overall returns, but decrease our volatility, as we were making money when the market moved in either direction. We created what appeared to be a virtually risk-free model based on the 75 quarters of data, and posted a return of over 1300% in this period; far more than you’d get by simply putting your savings in the bank. In theory, with the appropriate technology and start up money, you could run a programme where a computer managed your money and executed trades for you on the publication of new data, so you could sit back, relax, and enjoy the spoils!
After submitting a report, complete with analysis, we were assessed by six independent panellists in partnership with Georgia State University. Finishing in the top five, we then had to turn our report into a presentation, which we would present for eight minutes and then take questions for a further seven. We did this, and at the drinks reception afterwards found that we had achieved a very respectable second place.
Aside from the professional element, I managed to spend some time getting to see parts of the city and American culture. Donning the PPE hat, I visited the Jimmy Carter Presidential Library and Museum, as well as seeing a Major League Baseball (MLB) game, and sampling the ‘bigger is better’ food on the American menus. Being my first trip to the United States, I’d like to take this opportunity to thank Philosophy staff at the University of York for pledging money to support me. The experience will give me something to talk about in upcoming interviews when I look to apply for Graduate Programmes in Finance at the start of the next academic year, and will really help drive my case for converting my Summer Internship in Trading into a full-time position, due to the applied and unique knowledge and experience that I’ve gained.
Theo Wilson (PPE)
Tom Armstrong (Mathematics)
Edward Bottomley (Economics and Finance)
Jasmine Gotobed (Economics)
Gabriel Zedda James (Economics and Mathematics)
Leave a Reply