Tuesday, January 25, 2011
Data Distributions: The Gaussian Copula & Fat Tails
For a basic explanation of mortgage backed securities & toxic assets as they relate to the credit crisis see:
The Credit Crisis Visualized Part 1
The Credit Crisis Visualized Part 2
From: In defense of the Gaussian copula, The Economist
"The Gaussian copula provided a convenient way to describe a relationship that held under particular conditions. But it was fed data that reflected a period when housing prices were not correlated to the extent that they turned out to be when the housing bubble popped."
Decisions about risk, leverage, and asset prices would very likely become more correlated in an environment of centrally planned interest rates than under 'normal' conditions.
See also: Models and Agents-Hit by a Fat Tail
"Economists talk about fat tails when they want to refer to the probability that extreme events such as the above occur in a much higher frequency than you would regard as "normal."
"So why tails? And why fat?!" read more to find out.
Some examples of data generated from various copula functions and assumptions (using R) :