Fri, 15/08/2003 - 08:00
Professor Raphaël Douady, research director at Riskdata, outlines a new methodology for analysing hedge fund returns and measuring portfolio risk profiles. It is the first system which allows the calculation of a fund of hedge funds portfolio's aggregated risk, leveraging all available information on the underlying funds.
Managers of hedge funds and funds of funds are under increasing pressure from end investors to improve risk transparency and demonstrate the extent of their risk mitigation.
Whilst risk management technologies are widespread, to date they do not enable investors to quantify hedge funds 'relative risk profile' - i.e. relative to all the other assets as well as to each other - in order to evaluate their own aggregate risk situation.
Previous approaches to this issue based on 'positional transparency' have raised practical problems, such as the confidentiality of business sensitive information and the cost of simulating complex instruments.
In addition, positional-based risk profiles are biased as they do not take into account the dynamic behaviour of the portfolio resulting from the way a manager interacts with market events.
New technology from Riskdata addresses these core issues by allowing risk to be aggregated across all styles, based on a thorough and correlation analysis with a very large number of risk factors. This is made possible by enabling the following:
To evaluate the risk profile of a hedge fund purely based on its current market exposure would be like assessing the risk of an equity stock by only considering the company balance sheet.
In fact, a number of factors such as the company strategy, the management team and the situation in the connected markets will have as much impact, if not more, on the risk profile of a stock.
Similarly for hedge funds, as an investor what you buy is not its current market exposure, rather you are buying an investment style and the capability of a fund manager to produce a good risk/return profile.
This methodology therefore provides an analysis of each individual hedge fund, however most importantly it also provides the "relative risk situation" of the fund within a set of other hedge funds (e.g. for a fund of funds) as well as with respect to the rest of the market.
It is based on a thorough and accurate correlation analysis with a very large number of risk factors (over 15,000), then on the identification of the 5-10 risk factors that most matter in the fund risk.
With this information it is immediately possible to:
Riskdata's research team is currently running a research program, based on this technology, in connection with some leading fund of funds and hedge funds, and specialists in finance and mathematics disciplines. The results of this program will be presented in the last quarter 2003.
This approach was presented at a recent PRMIA/QWAFAFEW event in New-york to an audience of fund managers, funds of funds, CTA, risk managers, regulators and consultants. The technology is currently being tested by some leading funds of funds to support their risk management, their research and their allocation.
Riskdata's methodology makes use of all available information, including:
Background Note: Riskdata offers a comprehensive suite of solutions for Alternative Management, adapted to all investment styles: E.g. Fund of Funds, Long Short Equity, Fixed Income Arbitrage, Convertible and Volatility Arbitrage etc. Riskdata's team includes Investment Practitioners, Risk Management Experts and I.T Specialists. Its aim is to offer to all money managers easy, interactive and intuitive access to a powerful unified risk framework. It is supported by leading figures, such as Professor Robert Mundell, a past winner of the Nobel Prize for Economics. It is the first service offering a daily updated view on all market classes: equities, bonds, listed and OTC derivatives. As an interactive system, rather than classic ASP model, there is no exporting of clients' positions while deployment is quick and simple.
Copyright Hedgeweek 2003
Wed 21/01/2015 - 10:22
Thu 28/01/2010 - 11:54
Fri 05/12/2008 - 06:05
Tue 25/11/2008 - 06:01
Wed 21/01/2015 - 10:22
Thu 22/05/2014 - 12:01
Thu 24/04/2014 - 06:04
Fri 28/03/2014 - 10:35
Wed, 25/Nov/2015 - 15:30
Wed, 25/Nov/2015 - 09:20
Wed, 25/Nov/2015 - 09:16
Wed, 25/Nov/2015 - 09:14
Wed, 25/Nov/2015 - 08:28
Tue, 24/Nov/2015 - 16:33