Skip to main content

in this section

Financial Engineering Masters students complete a variety of challenging projects for Wall Street clients

Tuesday, March 25, 2014

Students at the December 2013 graduation ceremony, where they received Financial Engineering Concentration certificates.

Last fall, ORIE’s Financial Engineering concentration students devoted their third and final semester to team-based projects for real clients, including Alliance Bernstein, Banco Bilbao Vizcaya Argentaria, Cantor Fitzgerald, Citibank, Guggenheim Partners, KALX, and HETCO.  At ORIE’s Cornell Financial Engineering Manhattan (CFEM) offices near the New York Stock Exchange, they brought the insights acquired earlier in their Ithaca classrooms to an array of problems posed by clients.

Creating a portfolio of companies likely to be acquired

One team was sponsored by ORIE alumnus Timothy Tien ’88 of Guggenheim Partners (seen on the far right of the team photo below).  With faculty advisor Associate Professor Pierre Patie (who recently joined ORIE’s faculty), the team acquired (on paper) a portfolio made up of the stock of companies likely - according to the team’s detailed analysis - to be acquired in the near term.  Since only a small portion of the companies might actually be acquired during the timeframe under consideration, they developed a way to hedge the portfolioGuggenheim Team to limit their exposure to risk.  In financial engineering terms, work that increases the reward for a given level of risk is known as “finding alpha.” 

Several firms identified by the student team as ripe for acquisition were in fact acquired while the project was underway, yielding significant benefits to the portfolio.   According to Tien, the hedged portfolio would have earned 30% on an annualized basis had the shares actually been acquired and hedged.

Tien said, “I was so pleased that the team and our advisor were so intent on finding alpha and hedging the portfolio effectively.  We were able to successfully build on top of the groundbreaking work of the 2011 and 2012 teams, and take things to a level where we could conceivably run this strategy alongside other strategies in a systematic hedge fund or statistical arbitrage prop desk, and employ leverage.”

Simulating the operations of an equity derivative trading desk

Another group of financial engineering students built software to simulate the operations of an equity derivative trading desk, including the analytical tools needed to support the decisions traders make. Using Java and Excel, they built a “production system infrastructure” that allowed them to keep track of the trading activities and risk measures needed to carry out simulated trades of derivatives –put and call options – and the stocks that underlie them.   

The group then split into two teams that used the tools to simulate the buying and selling of stocks and derivatives.  Although BBVA Teamneither team did more than break even over the course of the study, they acquired substantial learning without financial or employment risk. The group seen in the photo at right worked with  ORIE M.Eng. alum Peter Robert ’79 M.Eng. ’81 (on the left) at Banco Bilbao Vizcaya Argentaria (BBVA) and faculty advisor Sasha Stoikov(second from the right), Senior Research Associate at CFEM. 

Stoikov recalls that "once or twice a week, Peter and I called the two teams to get price quotes on various exotic financial instruments. The team with the best price got the trade.  The students learned very quickly that it's not just about quoting the best price; you also need to understand how to hedge in order to be profitable."

Team member William (Dunfei) Chen, who has a job at a private real estate fund, recalls that "this was a real project that gave me hands-on experience in a subject in which I had some interest: I simply wanted to know more about derivatives and to dig deeper into the market-maker industry.  I would say that my expectations were perfectly fulfilled." 

Modeling the volatility of oil prices

The price of crude oil varies significantly over time as supply and demand react to changes in the industry, economic growth, geopolitical events and the like.  To counteract this volatility in prices, buyers and sellers are able to acquire the right to buy or sell oil at an agreed-upon price at a specified future time by buying or selling a “futures contract” at an options exchange.   The market for such contracts also affords a way to participate in the crude oil market without actually making or taking delivery of the oil itself.  This in turn has led to the creation of mathematical models with which a price can be determined for an option contract.   Not surprisingly, most such models incorporate estimating the volatility of oil prices, which makes the study of price volatility an important activity for financial engineers. 

Working with client Jean Payan of energy trading firm HETCO (jointly owned by Hess Corporation and a group of trading professionals) a team of ORIE financial engineering students used optimization techniques to analyze different approaches to modelling oil price volatility.  Among these was a proprietary approach that they could approach by treating it as a “black box.” That made it difficult to estimate model parameters using optimization techniques that rely on knowledge of the model’s formulas and which depend on the model having certain ‘nice’ properties.   

Nonetheless the team was able to use heuristic optimization methods to gain insights into the estimation of the parameters and to propose an improved methodology that might be pursued by a financial engineering team in the future.   

The team was advised by ORIE Professor James Renegar, who said that “the team members were phenomenal in their concerted effort: everyone was fully committed to advancing the project as much as possible in one semester."

Market microstructure and algorithmic trading

In addition to the trading simulation project, Stoikov advised two other project teams that studied the microstructure of equity markets.

One, with client company Cantor Fitzgerald, examined how brokers can use high frequency data to improve their business. The team used an optimization method to estimate how trading is distributed among different exchanges and built a model to make short term price predictions and improve the return from the liquidation or purchase of shares.

Another project advised by Stoikov, for Alliance Bernstein, investigated how the price of a stock is affected by activity in the options markets. The team determined that the net option market maker’s position has a strong negative correlation with the concurrent stock price but a strong positive correlation 20 minutes later. The students backtested two trading strategies to illustrate how the prediction models can be used in practice.

"In these projects," Stoikov said, "students learned what 'big data' in finance really means.  They were able to extract insights from high-frequency data sets that don't fit on your average laptop." 

“Hedge when you can, not when you have to”

ORIE Assistant Professor Andreea Minca advised a project team which investigated the effectiveness of different hedging strategies for using purchase and sale of a stock to replicate the payoff of so-called European call options written on that stock.  For consulting firm KALX, founded by mathematician Keith A. Lewis, the team built a simulation in C++, with an Excel add-in to analyze profit and loss, that they used to investigate different hedging strategies and to develop a proposal for an improved strategy.  The team titled their project report “Price-based hedging: Hedge When You Can, Not When You Have To.” 

Modeling term structure under equilibrium assumptions

A team sponsored by Citigroup and under the guidance of Visiting Assistant Professor Tibor Janosi investigated the interest rates that are paid to holders of mortgage position as a function of the length of time until they mature - the so-called term structure.  An effective means of forecasting term structure is a prerequisite for projecting investment returns and hedging portfolios against risk. 

One class of models to predict the term structure is based on the assumption that interest rates tend toward a long-term “normal” level but short-term economic effects ("noise") can and do prevent these levels from being achieved.  Such assumptions of tendency towards equilibrium are frequently used in economic theory.  The M.Eng. financial engineering team tested and compared several such models, which required calibrating and simulating each tested model.  Recent ORIE financial engineering graduate Weixuan (Joshua) Jiang  M.Eng. ’13 at Citigroup assisted the team.                 

The team worked with Andrew Feigenberg and Sara Angrist of Citigroup.  “Working for Andy and Sara was an absolute pleasure,” said Shelton (Fengyi) Xie, who is now on Morgan Stanley’s market risk research team.  “They were specific in what they wanted, yet very tolerant about our time frame. The project motivated me to do research on various fixed income products as they are all driven by the term structure of interest rates.  I was able to apply what I learned from my classes in the MEng program in a more practical context,” he said, and “the project has acclimated me to what I am doing in my current work.”

back to listing