The most important trend in hedge funds is that they are pursuing and attracting institutional money, especially corporate pension and public pension funds. The Bank of New York predicts that institutional capital invested in hedge funds will increase from $60 billion to $300 billion in the next five years and will soon account for more than 50 percent of annual hedge fund net inflows.

This will bring significant changes in an industry where the capital has tended to come from wealthy individuals.

BNY expects that much of the money will flow to “fund of fund” managers since this provides investors with diversification while allowing them to get exposure to relatively small, specialized funds.

To attract more institutional funds – which hedge funds will have to do if they are to grow rapidly – funds will have to become more structured in their approach and methodology. This also applies to investment advisors who are now looking to transition to a hedge fund model. Three of the most important factors that institutional fund managers will expect to see before placing money with a hedge fund include the following:

• An innovative investment strategy backed by a strong quantitative methodology
• Excellent market and credit risk management
• A sophisticated client interface and a robust computing architecture

While many hedge funds and investment advisors have deep experience in specialized areas, they have not consistently supported their strategy with state of the art quantitative finance techniques. Through a number of consulting engagements, CTC-Manhattan has helped clients in the development and vetting of complex analytics required to underpin and position their strategy. Some of the areas in which we have built applications include the construction of volatility surfaces, the pricing of exotic derivatives, different types of bonds, and new insurance proof of concepts that have embedded option-like features.

CTC is also a leader in developing new approaches in the field of risk measurement and management. We have worked in a number of areas, particularly bonds: callables, risky bonds, and convertibles.

CTC helps traders and portfolio managers get a picture of what might happen to their positions under a wide range of future scenarios. In most of our applications, Monte Carlo simulation is used to create many thousands of future scenarios for a range of market and credit conditions. Each of these future scenarios implies a future price for the instruments under consideration. Once these prices are computed, key aggregate properties (such as their distribution, VaR, CVaR, and worst case) are available for input into the decision-making process of the trader or manager. The process is iterative, in that the client can easily construct new sets of future scenarios for additional analysis.

Of course, risk management for complex portfolios has always demanded powerful computing. Sophisticated analytics requiring thousands of simulations per instrument can grind your infrastructure to a halt. CTC has deep experience in using industry standard components (MS, Intel) to build clusters and grids. CTC has helped develop low cost systems for its clients that run compute intensive risk analytics fast enough to meet the challenges of a competitive business environment. This technology has the added benefit of transparency – the trader or analyst simply uses their usual Excel front-end while Web services and scheduling software under the hood run their models using distributed computing resources on demand.

Perhaps the best way to understand one of the ways in which we deliver computationally intensive analytics to the desktop is to try one of our demonstrations – simply go to http://www.tc.cornell.edu/ctc-manhattan/callablebonddemo/ and follow the short set of instructions.

Conclusion

Hedge funds are different animals than mutual funds, of course: measuring and managing their risks requires the development of new analytical and computational models. Our team of financial engineers, applied mathematicians, and computer scientists is actively involved in this area. As we develop these new models, they will be tested and implemented using the computational platform (front-end Excel + .NET Web services + back-end commodity clusters) we have developed over the past several years.

For more information on work CTC is doing with hedge funds, please contact Roger Lang rlang@tc.cornell.edu.