Modeling the Mortgage market; it is all in the stars.

Around the time of Christ, Ptolemy the famous astronomer came up with equations and tables to explain the movement of the stars and planets in the night sky. The paradigm being that the earth was the center of the universe and everything else in the night sky revolved around us. His models worked well over the short term but over the long term aberrations occurred. Planets weren't in the exact part of the sky that was predicted but nearby. The modeling computed was a fair approximation and served its need for hundreds of years until the time of Galileo, Copernicus and Kepler when the earth was shown not to be the center of the universe and accurate models could be produced.
Fast forward to modern day advanced mathematics and computer modeling. Computers crunching statistics and advanced mathematics using probabilities come up with explanations and models of market observations. Instead of looking up to the stars, bankers are looking at reams of foreclosure data and debt defaults. Once the data fits the model well enough some brilliant person with a degree in finance figures out how to monetize it in the form of a derivative or rating a bond. Mortgage back securities are rated by bond rating agencies, credit default swaps are priced on over the counter markets and trillions of dollars are underwritten. The only difference now is that unlike the inertia of the universe in the night sky, markets are much easier alter and manipulate. Since the models and equations used become well known it is easy to fit the data for the wanted outcome. If you are are securitizing mortgages you can adjust what you put in to get the highest rating possible. Offsetting toxic mortgages with prime the system can be gamed and profits for the underwriters can be maximized. And if surgical cherry picking wasn't enough to corrupt the market; mathematical models only account for the known unknowns not the unknown unknowns. You set the model to see what the default rate is if we had a housing decline in line with the worst since the great depression and you kick out a value for the security. The models never account for what has never been seen before. How can they. It would be impossible to model all possible contingencies and would require greater imaginations than Hollywood screenwriters.
Credit default swaps are the same way. They "insure" by some estimates $45 trillion in debts. Again models are imperfect. Prices are determined by models and an over the counter market. While they are better indicators of the safety of the underlying debt than the bond rating agencies, they don't factor in a "perfect storm" scenario. Uncle Sam was ultimate guarantor of AIG swaps to the tune of $150 billion so far. A bankruptcy of could have caused a domino effect in debt markets that reverberated around the world possibly creating more Icelands.

The lesson should be to leave complex modeling to NASA and not economics. Instead we are insistent with regulating systemic risk and taking another slug of the hair of the dog that bit us. Financial derivatives may be great for a doctoral thesis in finance but when Wall Street gets ahold of them they become a weapon of worldwide financial destruction.