Tyranny of Averages
The tyranny of averages describes the often overlooked fact that the mean does not provide any information about the shape or skewness of a probability distribution.
- Averages can often hide as much as they reveal
- A typical measurement for the members of a dataset may not look like the average at all
- Also need to look at a distribution's shape and skewness
- Degree Distribution can help understand if average is a reflective statistic
- Average works well when data samples are clustered around the mean
- Average does not work well if there are outliers that skew the statistic
Illustrative Example of The Tyranny of Averages
- Assume 9 people with no savings are in a bar
- Their average net worth is $0
- In walks a billionaire with $10 Billions in savings
- The average of the people in the bar is now worth $1 Billion
- In reality though, no one in the bar only has $1 Billion or close to it.
- The single billionaire threw off the average
- The median savings of $0 would be a more accurate reflection
Real World Investing Implications
- Be careful of economic statistics that use averages - they may be skewed
- Ex. Income inequality. The super rich may be driving up average incomes, but in reality most people are getting pooer