# Thread: Statistics and Probability Risks and Gambles: 2008 housing market collapse

1. ## Statistics and Probability Risks and Gambles: 2008 housing market collapse

Say for example most of the people of the world were good in probability and statistics in the year 2008-2009.

My question is hypothetical.

Would we have been in the financial meltdown in the year 2008-2009 which was due to housing market collapse because of the direct result of the overwhelming sub-prime mortgage and foreclosures?

At that time hundreds of thousands of houses were foreclosing per month.

The small lenders were lending the money at a level that was beyond the limit of a normal interest rate. They were lending money who had a bad credit score.

If all the bank employees and administrators who dealt with mortgages were more strict and aware of risks and probabilities and combinations and permutations do you think we would be in the crisis that we were in at that time?

I am asking this because I came across this video by Ted(3 minutes) where Arthur Benjamin says that statistics and probability should be the apex of the all the branches of Mathematical pyramid and Calculus should be next to it.

2. ## Re: Statistics and Probability Risks and Gambles: 2008 housing market collapse

The genesis of the sub-prime mortgage collapse was directly caused by experts in probability and statistics, who under-estimated the risk of not knowing what they didn't know. They felt that by packaging sub-prime loans together the risk of a few loans going into default would be mitigated. Fine in theory, but it didn't account how defaulting on loans may not be strictly independent events. Nor did they take into proper account how changes in sub-prime lending practice would effect the statistical probability of individual defaults. As an analogy: if the probability of it raining on any given day is 10%, what's the probability of it raining 10 days in a row? Hint, the answer is not 10^-10; it's actually much greater. Same with mortgage defaults - if one person defaults the probability of the next guy also defaulting is low; but if 100 people default the probability of the 101st person defaulting becomes large, and you get a snow ball effect. In conclusion - you may think you know what to measure and how to calculate probability, but financial systems are not static, so whatever you measured using data from last week may not be valid for how the system will be operating next week.