Tuesday, February 24, 2009

The Financial Model that Broke the System

Lots of investors, journalists, businesspeople, and politicians have mentioned the fact that the financial industry's models failed to appropriately price the risk in the system and that the failure to do so contributed hugely to the current credit crisis.

However, that failure was not any single individual's fault. Wall Street's models were designed to do just that, to model the real world, using a set of assumptions that may or may not hold up in the face of reality. Somehow, though, these models became touted as manna from the financial gods, feeding the frenzy of "sophisticated" investment and risk-management strategies.

Until today, it had never occurred to me to look more closely at the actual models used by Wall Street. Why would any non-mathematician intentionally attempt to decipher a formula like this? (Source: wired.com)


It looks ridiculous, right? What on earth is this mess of Greek characters supposed to mean? Perhaps that hesitation to peer through the looking glass is what caused investors to miss the shortcomings inherent in their models to see the real risks they faced.

In a new article for Wired Magazine, Felix Salmon tells the story of that formula, David X. Lee's "Gaussian Copula Function," and makes it not only understandable to a layperson, but also as entertaining as an article about a mathematical model can ever be. I highly recommend it to anyone interested in learning more about what went wrong and how we arrived at our current economic crisis.

At the end of the day, John Maynard Keynes' wisdom on economic models lives on. "It is better to be roughly right than precisely wrong." Many people on Wall Street forgot that essential edict and now we are all paying the price.

No comments: