Overview

A lognormal variable is one whose logarithm is normally distributed. It suits positive, right-skewed data such as incomes, house prices and file sizes.

Enter x (above 0), and μ and σ, the mean and standard deviation of ln X.

Worked examples

InputsResult
x (>0)20Mu (μ of ln X)2.8Sigma (σ of ln X)0.55Lognormal PDF: 0.03CDF: 0.64Median: 16.44Mean: 19.13
Show 2 more examples
x (>0)1Mu (μ of ln X)0Sigma (σ of ln X)1Lognormal PDF: 0.4CDF: 0.5Median: 1Mean: 1.65
x (>0)50Mu (μ of ln X)3Sigma (σ of ln X)0.4Lognormal PDF: 0CDF: 0.99Median: 20.09Mean: 21.76

Lognormal PDF by x

With the calculator's starting values, the lognormal PDF is 0.03. The table shows how it changes with the x.

x (>0)Lognormal PDF
100.05
12.50.05
150.05
17.50.04
20starting value0.03
22.50.03
250.02
27.50.02
300.01

Other inputs: Mu: 2.8 μ of ln X · Sigma: 0.55 σ of ln X. Change them in the calculator above for your own numbers.

How it works

The CDF is the normal CDF of (ln x − μ) ÷ σ. The median is e^μ, and the mean is e^(μ + σ²/2), which is always bigger than the median because of the long right tail.

The formula

P(X ≤ x) = Φ((ln x − μ) ÷ σ)
μ, σMean and standard deviation of ln X, not of X itself

Practical tips

  • With μ = 2.8 and σ = 0.55, P(X ≤ 20) is about 0.64, the median is 16.44 and the mean 19.13.
  • To estimate μ and σ, take the log of your data and find its mean and SD.
  • The bigger σ is, the bigger the gap between the mean and the median.

Frequently asked questions

Why is the mean above the median?

The right tail pulls the mean up.

Can x be zero or negative?

No, lognormal values are always positive.

Where is it used?

Finance, biology, reliability and many measurements that grow by percentages.