Kurtosis indicates whether a distribution is cusped or smoothed compared to the normal distribution. Index of kurtosis (E) is calculated using the following formula:
,
where:
s. Standard deviation of the sample.
n. Sample size.
xj. j-th element of the sample.
. Average.
If E < 0, data is distributed along the entire range of values more evenly , if E > 0, data is concentrated around the arithmetical mean.
See also: