Research by Russell Wermers
Hedge fund monitoring lag solved; breakthrough formula offers a new shield for institutional investments
In August 2007, large hedge funds began suffering significant losses mid-month. Investors were left guessing about both the performance of their managers and the impact of losses on their overall portfolios.
Such volatility, compounded by extreme market movements and uncertainty heading into 2008, fueled a recently published, award-winning study by Associate Professor of Finance Russ Wermers and co-authors Michael Marcov and Daniel Li.
Scope of the Crisis
Hedge funds provide minimal, if any, transparency in terms of day-to-day and week-to-week performance and comprise roughly two trillion investment dollars rippling through the global economy – from wealthy investor holdings to pension funds, and nonprofit foundations and endowments.
“It’s important for these entities to know their funds’ rate of return between report dates, which typically are monthly, or quarterly in some cases,” Wermers says. “This time lag generally has been OK. But in the wake of our global financial crisis – or even when market volatility is abnormally high – you want to know what’s going on with your fund on a daily or at least weekly basis. Right now, it’s unacceptable to not know whether you’re too deep into a risky pool of assets.”
Wermers’ answer is grounded in synthetic replication portfolios based on a monthly factor model using common investable indices.
The formula “uses monthly returns on more than 100 indexes (e.g., broad stock, bond, currency and commodity) to fit the monthly returns of a particular hedge fund, then forecasts the daily returns of that hedge fund during the following month using the publicly observed daily returns on the indexes that best fit the hedge fund,” explains Wermers. “This can both control for daily hedge fund risk and estimate and control value-at-risk.”
Spurring the breakthrough is dynamic style analysis (DSA), a technique developed by Markov. This time-varying linear regression method improves upon traditional moving window models by using flexible least squares methodology, which allows risk exposures to change over time.
“DSA provides an approach to select best-fit factors through a pool of more than 100 market indices and factors,” says Wermers. “Once a high quality in-sample replication is achieved, we use daily returns of replicating assets reported in the financial press to create a daily proxy of the hedge fund.”
“The trick here lies in mirroring the real individual hedge funds with combinations of the hedge fund indexes,” he adds. “For example, take 20 percent of market index one, 40 percent of market index two, and 40 percent of market index three, then it’s easy to look up those three indexes each day, apply those three weights to come up with a daily simulated return for your fund.”
The application is designed to be user-friendly for investment managers, advisors, consultants – with at least basic skills in running regressions – “and even family offices that have a bit of horsepower,” he says. “And, Markov Processes International provides software and services to implement this approach.”
Moreover, the new model can leverage insight to mitigate negative externalities imposed by actions of other investors, such as those who are first to exit a fund during a liquidity crisis.
“By the time an investor’s told ‘you’ve lost 20 percent at the end of the month,’ it’s too late to do anything,” Wermers says. “In such cases, it’s best to know when the loss is at 5 percent, so you can get out or at least get the process rolling before the fund loses, let’s say, 50 percent of its value.”
Published by the Journal of Investment Consulting, “Monitoring Daily Hedge Fund Performance When Only Monthly Data is Available” earned first place in the Investment Management Consultants Association’s 2012 Academic Paper Competition and coincided with Wermers’ selection to help rewrite the criteria for the examinations for asset management professionals wishing to achieve the prestigious CIPM Program designation of the CFA Institute. For more information, contact firstname.lastname@example.org.
|Previous Article||Table of Contents||Next Article|