08 October 2008

Conditional probability is subtle

From 60 Minutes on Sunday, video here: the banking crisis is the fault of the mathematicians and physicists, who went to work on Wall Street and invented some complicated models. The "financial expert" says that "you can't model human behavior with math". That may be true with our current mathematics, but I suspect the fault doesn't lie with the modelers so much as with the people who went ahead and thought the models were perfect.

Although there are rumors that a lot of the models basically worked on the principle that defaults on mortgages were independent, which is so obviously false that you'd fail a freshman who said it. (Basically, when the economy gets bad, more people can't pay their mortgages. The effect is larger because there are adjustable-rate mortgages, and everybody suffers roughly the same adjustments.) I would not be surprised to learn that the problem here is that Conditional Probability Is Subtle; the quants may very well have known their models were flawed, but the suits didn't want to hear it.

I just hope this meme dies out quickly and doesn't gain traction. Our PR already isn't so good.


Aaron said...

You can model human behavior pretty well with math if the human in question has fallen off a cliff. ;)

On a more serious note, here's a related article in the NYT, in case you haven't seen it.

Blake Stacey said...

It's actually a Very Elaborate Conspiracy to hold the financial industry hostage: fund ITER and the LHC, or our comrades who "dropped out of grad school" will destroy Wall Street. Moo hoo ha ha!

misha said...

According to the 10.03.2008 episode of This American Life the real problem is that so many of the bad mortgages were used to underwrite many financial derivatives. The credit market froze when it turned out that these derivatives were practically worthless. See How the markets really work for a more humorous perspective, explained by two British comedians.

Anonymous said...

I (surprisingly) more or less agree with misha. The problem has nothing to do with either the models or the modellers. The problem is in the lack of regulation and transparency.

Part of the input into the model for any given CDS is "what's the probability that the other guy will fail?" But when he's also trading CDSes left and right, and when he doesn't have to disclose what he owes to whom for those CDSes, then I have no way of determining that probability other than a blind guess.

People used to make those guesses and ran with it. But when the mortgages crumbled they caused chains of these debts to break, which exposed the problem. Now people know that they don't know this information, and instead of making that guess they refuse to take any chances at all. Thus the markets freeze over.

Paul Soldera said...

How about 'Algorithms don't kill people, people kill people' as a slogan? It's a bit more hard hitting than 'Conditional Probability is Subtle'.

PR is all about assertiveness.

misha said...
This comment has been removed by the author.
misha said...

I would not say that the crisis has nothing to do with the models or modelers. Let's remember that the purpose of these models is to maximize the amount of money that you can siphon into your pocket from the fluctuations of the market. Let's be earnest, the massive computerized speculation adds significantly to the instability, and keeping the credit default swaps private and secret makes things much worse. I'm pretty sure that the speculators are making out like bandits on the current wild volatility. Let's hope that they will invest their windfalls into something worthy. As for the housing market, the prices collapse could be a good thing, it would bring the hyper-inflated housing prises to a more reasonable level, and everybody would be better off. Unfortunately these prices got inextricably interconnected with the credit market, investment banking, and much more in our financial system. Let's hope that the investment banking will get again separated from the consumer banking, as it was before the final stage of deregulation.