Each year the Chicago Quantitative Alliance (CQA) holds a portfolio management contest for university students, where the students must manage a portfolio with strict requirements. This past year, the contest ran from the end of October until the start of April. How did the students do? What did their portfolios look like?
The CQA is a non-profit, professional investment organization composed of quantitative investment practitioners. I'm a member. The CQA’s primary goal is to “promote the interests of the quantitative investment community” and it is composed of academics, consultants, investment managers and other professionals. Although the group started in Chicago in 1993, only 15% of the members come from Chicago, and the majority of the membership comes from the East coast.
The students were tasked with creating portfolios on StockTrak and they were provided access to CQA mentors, who guided them along the way. The student groups were also given access to Markit, Axioma and FactSet. When creating the portfolios, the students had to follow certain rules:
- The portfolio had to have a beta of +/- 0.5
- The portfolio had to be long/short portfolios that were market neutral.
- The ‘universe’ of potential stocks that they could choose from, was limited to 1,000 liquid large and midcap stocks.
- The portfolio had to have less than 5% of its holdings in cash.
If we were to put all 34 portfolios together into one pool and analyze them, what would they look like? I reached out to Chris Martin at Axioma and he noted that:
The majority of Factor Risk came from Style Factor tilts, or factors based off of market traded descriptors (i.e. Exchange Rate Sensitivity, Medium Term Momentum and Volatility) and balance sheet descriptors (Value, Growth and Leverage).
Martin added:
In looking at the Style Factor exposures through time, we see that the Fund had a constant positive exposure to Medium-Term Momentum, and a constant negative exposure to Market Sensitivity – or “Beta”.
We can get an idea that students were choosing momentum stocks, but how did that strategy perform? By the end of the competition, the students had an average return of -2.05%. But, that’s only part of the story and it turns out that throughout most of the competition, the students were beating the market. From the end of October, 2015 until February 11th, 2016 the Dow fell -11.3% and then the Dow experienced a dramatic upswing as it rose 13.6%. This whipsaw and volatility in the market may have caught some students (and other mutual and hedge funds) off-guard. From October 30th, 2015 until April 1st, 2016 the Russell 1000 was up 0.3% and the students had an average return of -2%.
Martin, elaborated more on their performance:
In digging deeper, there were two main drivers of this dramatic drop in performance in only a short time period: the net short exposure to Market Sensitivity and the net long exposure to Medium-Term momentum. The net short exposure on Market Sensitivity paid off for the initial several months, and in fact this seemed to cause even larger short bets on this factor. Unfortunately, the Market Sensitivity factor started performing quite well (namely, higher beta stocks started performing better than low beta stocks). The net long exposure on Medium-Term Momentum was consistent through time, but never drove any positive performance – only volatility. Towards the end of the time period, Medium-Term Momentum started performing poorly, which greatly detracted from the portfolio’s performance.
Student groups were judged not only on their returns, but also on their risk-adjusted returns and a video presentation that they had to make for the competition.
Ileana Cavazos, Jielin Chen, Mauro Hamz, Armin Niorumand, and Anton Vlaykov from The Stuart School of Business won the competition. The team’s CQA mentor was Binoop Unni from Markit, and their advisor was John Bilson, Associate Dean and Professor of Finance at Stuart.
To create their portfolio, the team combined 3 quant strategies in choosing their stocks. The team combined Greenblatt’s ‘magic formula’, Piotroski’s financial health score and Mohanram’s G-Score. Greenblatt’s ‘magic formula’ looks at the EBIT/EV and Return on Capital for a stock. Piotroski’s financial health score is a financial health score that looks at 9 factors and assigns a score from 0 to 9 on a stock. Mohanram’s G-Score is another multi-factor score that takes into account 8 factors based on profitability, expenditures and growth. In some manner, the team essentially made a 19 factor screen by using only 3 factors.
The overall portfolio shorted energy and went long on consumer non-durables. That bet paid off, as the energy sector (using the XLE SPDR as a benchmark), fell -8.8% during the competition and according to their calculations it contributed to 14% of the return for their portfolio. For the group, their best performing stocks came from shorting
The team from Stuart had an absolute return of 19.9% in the competition. Here's a video of the winning stock portfolio:
