Definition:Excess Kurtosis

From ProofWiki
Jump to navigation Jump to search

Definition

Excess kurtosis is defined as the difference between the kurtosis of a particular probability distribution and that of the Gaussian distribution.


Let $X$ be a random variable with kurtosis $\alpha_4$.

By Kurtosis of Gaussian Distribution, the kurtosis of a Gaussian distribution is equal to $3$.

So the excess kurtosis of $X$, usually denoted $\gamma_2$, is given by:

$\gamma_2 = \alpha_4 - 3$


Also known as

Some sources give this as kurtosis excess.


Also see


Sources