MSCI Barra, a New York-based indexing and analytics firm, has launched the Barra Hedge Fund Model, which quantifies a hedge fund’s market risk and the risks associated with a fund’s strategy and style, in a portfolio context."There are a number of ways to measure hedge fund risk, but they are not always compatible with looking at risk of traditional investments. Our clients, particularly pension funds, were asking to give them an integrated way of looking at fund risk,” said Jonathan Hudacko, a product manager at MSCI Barra, whose clients have been beta-testing the model for about a year.
The hedge fund model obviates the need to access portfolio holdings information, by using step-wise regressions to determine the portion of fund’s returns explained by known factors, such as equity, fixed-income, currency and commodity exposures. It also accounts for time-dependent information, such as changes in geography concentrations and changes in trading strategies.
Hudacko added that for long/short equity hedge funds, most risks can be explained by equity factors. However, for convertible arbitrage, managed futures or distressed debt funds, their trading behaviour plays a key role in explaining risk exposures. This means that a more complex risk model than a simpler index-based approach is needed.
However, the model has certain limitations, due to the lack of transparency into most hedge funds. “The investor typically doesn’t have transparency into the portfolio, so we don’t have clear information such as non-linearity in the portfolio, what positions the hedge funds have, or information about their liquidity. The model is designed primarily to pick up market and alternative systematic risks, but there are other risks outside of these that may become highly significant in times crisis," said Hudacko.
Sign up for Risk.net email alerts
Singapore, 22nd - 23rd Jul 2014
Australia, 12th - 13th Aug 2014
UK, 10th - 12th Sep 2014
There are no comments submitted yet. Do you have an interesting opinion? Then be the first to post a comment.