http://www.wired.com/techbiz/it/magazine/17-03/wp_quant
Talks about a financial mathmetician who developed an equation to quantify the risk of investment vehicles and how it fueled the boom and ultimately led to the meltdown.
Conspiracy theorists have some fuel too, as the guy is now under wraps back in China…
The role of quantitative analysis, particularly as it has been applied by analysts using AI software, is also discussed in the current issue of H+ magazine (see “The Global Financial Crisis…”) in http://hplusmagazine.com/...edition/2009-spring/). I also brought up the role of the quants near the end of the “How Will We Know Whether It Worked?” thread. The basic weakness in the mathematical approach is that it requires certain assumptions about randomness that may not be valid in the real world of human action. For instance, the assumption that you won’t have a mathematically improbable clustering of mortgage defaults may not apply if a political policy is in place encouraging lending to marginal buyers, or if interest rates are being manipulated exogenously.
A long but interesting VDO on the crisis . . . http://www.hulu.com/watch/59026/cnbc-originals-house-of-cards
Another interesting explanation…
…For instance, the assumption that you won’t have a mathematically improbable clustering of mortgage defaults may not apply if a political policy is in place encouraging lending to marginal buyers, or if interest rates are being manipulated exogenously.
Ah, but surely that could never happen, right? 
Conspiracy theorists have some fuel too, as the guy is now under wraps back in China…
That is interesting.
“Li has been notably absent from the current debate over the causes of the crash. In fact, he is no longer even in the US. Last year, he moved to Beijing to head up the risk-management department of China International Capital Corporation. In a recent conversation, he seemed reluctant to discuss his paper and said he couldn’t talk without permission from the PR department. In response to a subsequent request, CICC’s press office sent an email saying that Li was no longer doing the kind of work he did in his previous job and, therefore, would not be speaking to the media.”
**
and this…
"As Li himself said of his own model: “The most dangerous part is when people believe everything coming out of it.”
**
Here are a couple fun conspiracy theories:
One thing is for sure: Math is a science; banking and economics are not.
Oh no, I was thinking purely hypothetically.
In defense of Li, he did issue some early warnings about the limitations of his construct.
“One thing is for sure: Math is a science; banking and economics are not.”
I would say that economics can be a science, but not a quantitative one, at least if one is talking about cardinal (rather than ordinal) quantities. (Not all science has to involve precise quantitative measurement; biological taxonomy, for example, is based primarily on the recognition of differences rather than measurements of differences in degree.) Since economic action arises from individual value scales which can assign only ordinal rankings to alternatives, we can really only draw comparative conclusions about economic processes–which is why (mathematical) econometrics is IMO a misbegotten endeavor. Similar criticisms can be leveled against the sort of quantitative analysis being used in finance today.
I have spent most of my career doing statistical and mathematical models of complex physical systems. Before I read the Wired article I would have guessed that the model failed primarily because people manipulated it to show what they wanted to see.
After reading the Wired article it seems to me that clearly one aspect of the model failed, i.e. the reduction of a complex series of risk factors to a single number, and it failed because (1) it was not possible to do so accurately with any model and (2) the historical data used to “calibrate” the model was too short in duration to understand the range of possible outcomes. (and other reasons…the separation between the people who understood the model complexity, and the decision makers who understood the market etc…)
But…I also got the impression that the model(s) actually worked very well as they showed all of the people who were running them that the system was completely controlled by housing prices, and if the housing prices dropped the whole thing blew up. To me, that is a successful model as it highlighted the one critical factor in a complex system. It also confirms that many people in the financial industry knew quite clearly that if the housing prices were ever to drop the whole house of cards would collapse and that hiding behind the “nobody understood the complexity” issue is BS.
Gets back to an older post of mine that suggested these Folks should be charged under RICO as they had to have understood that if house prices dropped, the market would collapse, but then they would be too big to fail and could essentially blackmail the govt. into backstopping them.
Great article if you are any kind of math geek.
One thing is for sure: Math is a science; banking and economics are not.
I was on the phone today with a rep from my mortgage company. He dropped out of Industrial Engineering (referred to as “imaginary engineering” where I went to school), changing his major to finance. He works at Countrywide. 'nuff said.