Kurtosis

AdvancedRisk Management2 min read

Quick Definition

A statistical measure of the "tailedness" of a probability distribution, indicating how likely extreme outcomes are compared to a normal distribution.

What Is Kurtosis?

Kurtosis measures how heavy the tails of a distribution are relative to the normal distribution. Higher kurtosis means more extreme outcomes (both positive and negative) than expected.

Types of Kurtosis:

TypeExcess KurtosisMeaningExample
Mesokurtic= 0Normal distributionTheoretical benchmark
Leptokurtic> 0Fat tails, peaked centerStock returns (typical)
Platykurtic< 0Thin tails, flat centerUniformly distributed returns

Excess Kurtosis in Financial Markets:

Asset ClassTypical Excess Kurtosis
Large-cap stocks (daily)3 to 10
Small-cap stocks (daily)5 to 15
Currencies (daily)2 to 5
Government bonds (daily)1 to 4
Commodities (daily)3 to 8
Normal distribution0

Why It Matters for Investors:

  • High kurtosis = more frequent extreme moves than expected
  • Risk models assuming normal distribution underestimate true risk
  • Portfolio VaR calculations should be adjusted for kurtosis
  • Options tend to be underpriced when kurtosis is high

Practical Implication: If your portfolio's return distribution has excess kurtosis of 5, a "1 in 1,000" event under normal distribution actually occurs roughly 1 in 100 times — 10x more frequently.

Formula

Formula

Kurt[X] = E[(X-μ)⁴] / σ⁴ - 3 (excess kurtosis)

Kurtosis Example

  • 1S&P 500 daily returns have excess kurtosis of ~5 — meaning extreme moves happen far more than a normal distribution predicts
  • 2High kurtosis in crypto markets (15+) means 5%+ daily moves are common, not exceptional