What is Kurtosis?

Kurtosis is a statistical measure that describes the "tailedness" of a distribution — how prone it is to extreme values, or outliers, compared with a normal distribution. In finance, kurtosis is hugely important, because it helps explain why extreme market events happen more often than simple models predict. This guide explains what kurtosis is, the different types, how it relates to "fat tails", and why it matters — in clear, plain language. It builds on the normal distribution and is a core topic in risk qualifications like the FRM.

What is kurtosis?

Kurtosis measures the shape of a distribution's tails — specifically, how much of the data sits in the extremes (far from the average) versus near the centre. It's often loosely described as how "peaked" or "flat" a distribution is, but the more accurate and useful interpretation is about the tails: a distribution with high kurtosis has fatter tails, meaning extreme outcomes are more likely than in a normal distribution.

The four "moments" of a distribution

Kurtosis is best understood as one of four statistical "moments" that together describe the shape of a distribution, each building on the last:

• The mean (first moment) — the centre, or average.
• The variance / standard deviation (second moment) — the spread around the centre.
• Skewness (third moment) — the asymmetry, or lean, of the distribution.
• Kurtosis (fourth moment) — the tailedness, or propensity for extreme values.

So where the mean and standard deviation describe where a distribution sits and how wide it is, skewness and kurtosis describe its shape — and kurtosis specifically captures the extremes that matter so much for risk.

Kurtosis and the normal distribution

Kurtosis is measured relative to the normal distribution, which serves as the benchmark. The normal distribution has a kurtosis value of 3 (or, using the common "excess kurtosis" measure, a value of 0). Distributions are then classified by how they compare:

• Mesokurtic — the same tailedness as a normal distribution (excess kurtosis around 0).
• Leptokurtic — fatter tails and a sharper peak than normal (positive excess kurtosis). Extreme values are more likely.
• Platykurtic — thinner tails and a flatter shape than normal (negative excess kurtosis). Extreme values are less likely.

"Excess kurtosis" simply subtracts 3, so that the normal distribution sits at 0 and the sign immediately tells you whether the tails are fatter (positive) or thinner (negative) than normal.

Fat tails: why kurtosis matters in finance

The single most important application of kurtosis in finance concerns "fat tails". Real financial returns are famously leptokurtic — they have fatter tails than the normal distribution. In plain terms, this means that extreme market moves (large crashes and large surges) happen more often than a normal-distribution model would predict. This has profound consequences:

• Risk models that assume normally distributed returns — like the basic parametric form of Value at Risk — will underestimate the likelihood and size of extreme losses.
• The 2008 financial crisis is a stark example: events that "should" have been almost impossible under normal-distribution assumptions occurred anyway, precisely because real returns are fat-tailed.

Recognising and measuring kurtosis is therefore essential to honest risk management — it's a direct warning that the comforting bell curve understates the danger of rare, severe events.

Why it matters for finance professionals

For anyone in risk or quantitative finance, kurtosis is a vital concept. It captures a property of real markets — the prevalence of extreme events — that simpler models miss, and it explains why those models can be dangerously over-optimistic. Understanding kurtosis, fat tails and their implications is fundamental to sound risk modelling and a regularly examined topic in professional risk qualifications.

Frequently asked questions

What is kurtosis?

A statistical measure of the "tailedness" of a distribution — how prone it is to extreme values compared with a normal distribution. High kurtosis means fatter tails and more frequent extreme outcomes.

What is excess kurtosis?

Kurtosis measured relative to the normal distribution, calculated by subtracting 3. A normal distribution has excess kurtosis of 0; positive means fatter tails (leptokurtic), negative means thinner tails (platykurtic).

What are fat tails?

Tails that are heavier than a normal distribution's, meaning extreme outcomes occur more often than the bell curve predicts. Financial returns are typically fat-tailed (leptokurtic), which is why kurtosis matters so much in risk.

Why does kurtosis matter in finance?

Because real returns have fat tails, models assuming normal distributions underestimate the chance and size of extreme losses. Kurtosis flags this, which is crucial for realistic risk management — a lesson reinforced by the 2008 crisis.

Build your quant skills with Learnsignal

Kurtosis and fat tails are central to understanding financial risk. Learnsignal's tutor-led courses, including the FRM, develop the statistical and risk understanding that topics like this build on — with clear teaching that connects theory to why markets surprise us.