Risk providers respond (part two): What has the market crisis done to risk modeling?

This week on the podcast, we follow-up on our conversation with APT, Axioma, Northfield, Barra, and R-Squared. In the podcast, we discuss ways the different risk model providers have adapted before, during, and after the credit crisis. 

Subscribe to our podcast in iTunes to get each of the episodes in this series delivered directly to your computer. Can't get enough risk? Check out our Risk blog.

Some highlights from our episode:

Jason MacQueen, one of the founders of R-Squared

"Because market conditions have changed so rapidly, we've tried to make the model more responsive....Normally when we built a model we'd scale it to the volatility of a global portfolio over a one-year period. It's now scaled to a trailing six month ex-post volatility. That had the effect of making the model more reponsive to recent changes."

Dan diBartolomeo, President and founder of Northfield Information Services

"Starting around 2005...we had developed a very different way of looking at risk over shorter time horizons. Rather than look at high frequency data we chose to continue to use low frequency data but supplement the observed risks with information from alternative sources such as option markets and forms of volatility measurements such as Parkison volatility, cross sectional volatility...and other measures that are more observable in the short run."

Oleg Ruban, Senior Associate of Applied Research and Patrick D'Orey, Vice President, Head of Europe Equity Analytics of MSCI Barra

"Risk is not a single number, since the early part of [the 90s] we've been providing long and short versions of many of our models...we have a number of tools to help investors look into the tails of the return distribution..."

"We find that the market is now using more and more of these in conjunction [with one another]. Our approach to risk analysis hasn't changed that much."

Sebastian Ceria, CEO and Olivier D'Assier, Managing Director for the EU and Asian Markets of Axioma

"When Axioma embarked on this risk venture in 2005...we didn't believe in this religious discussion of fundamental models versus statistical models...[we] decided to provide multiple risk models as part of our platform...At the time it looked like this was not needed, it just so happened that when the crisis hit...it became a great competitive advantage."

Laurence Wormald, Head of Research, Sungard APT

"We have made two major efforts to think about improving the way our models work. One of them has to do with the way in which we look at global macro factors across regions and across asset classes."

"We've [also] undertaken a major effort to backtest and look at the outer sample performance of all our risk models. We're making the results of those backtests transparent to our clients."