Statistics

Kurtosis

A measure of tail heaviness, with the normal distribution as the reference point at 3.

Excess kurtosis subtracts 3, so a normal distribution has excess kurtosis 0. Positive excess means fatter tails than normal.

The trap. Kurtosis is often described as "peakedness". It is much better understood as tail weight - the fourth-moment weighting means extreme observations dominate the statistic almost entirely.

Consequently it is a terrible estimator in practice: it is extraordinarily sensitive to a handful of outliers, so sample kurtosis from limited data is unstable. For fat-tailed data the population value may not even exist.

Related terms

Practise this

Put it into practice

Knowing the definition is not the same as spotting where it applies under time pressure. Work the question bank free.

Start practising free

Browse the full quant interview glossary