Taleb Fat Tail
Nassim Taleb wrote about fat tail risks in the book Black Swan. Lets understand that by visually proving it. I took inspiration from this video of Nassim Taleb for this blog — Taleb on Fat Tails
Pre-requisites:
- Central Limit Theorem: No matter what the population distribution is, the distribution of sample means is going to be normally distributed.
Normal distribution has thin tails (and heads) — Extreme cases (outside 3 std) have only ~1% chance of happening.
Taleb is basically arguing that in certain situations, there will be fat tails — i.e the Central Limit Theorem Fails. What are these situations?
A population with huge inequality or variation.
If I consider entire world’s calorie consumption as my variable— the variance of it is still small. Central Limit Theorem applies well to this data and the distribution of sample mean of calorie consumption will be normally distributed.
But what if I consider percapita wealth as my variable? The population data is skewed — very less number of people own large wealth. As an example (from the video) I take a population of 1002 individuals. 1000 people earn uniformly between 5 Lakhs to 24 Lakhs. The other two earn between 50 Crores to 24 Crores. If this is the population, do you think CLT holds here?
Nope!
We see fat tails.
And when we apply models that rely on the normality assumption of the sampling distributions, we underestimate these risks.
That is why Nassim Taleb argues that Financial Risks are underestimated by the Wall Street. Because Financial markets can have fat tails given the huge variance possible in the population.
He says that Pandemics and Wars are two major areas where Fat tails exist.
Related blogs:
- Beware of Fat Tails {Pandemic}
