Log-normal distribution
A price is log-normal when its logarithm is normal. It can never go below zero and has a long right tail, so its mean sits above its median.
01The formula
- today's price
- the price T years from now
- drift: the expected return per year
- volatility per year
- years ahead
- a normal distribution with mean m and variance v
02See it move
03Where it is used
- Options deskPrices options by assuming the stock at expiry is log-normal, as Black–Scholes does.
- Portfolio managerShows clients the spread of future account values, which is skewed, not symmetric.
- Risk managerSimulates price scenarios that respect the zero floor.
04Why traders use it
Returns compound, so prices multiply rather than add. A normal model of prices allows negative prices and misses the skew; the log-normal gets both right.
05Worked example
A $100 stock with drift 8% and volatility 30%, 1.00 years ahead: the median price is $103.56 but the mean is $108.33. The chance it ends below where it started is 45.4%. A normal with the same mean and spread gives a 0.056% chance of a negative price.
06How it is built
- Model the log return over years as normal: .
- Exponentiate: , always positive.
- The median of is the exponential of the median log: .
- The mean adds back half the variance: .
07The common mistake
Treating the mean as the typical outcome. With high volatility most paths end below the mean, because a few very large outcomes pull it up.
08In Python and LaTeX
import numpy as np from scipy.stats import lognorm S0, mu, sigma, T = 100, 0.08, 0.3, 1.0 m, s = np.log(S0) + (mu - sigma**2/2)*T, sigma*np.sqrt(T) X = lognorm(s, scale=np.exp(m)) print(round(X.median(), 2), round(X.mean(), 2), round(X.cdf(S0), 4))
\ln S_T \sim N\!\left(\ln S_0 + (\mu - \tfrac12\sigma^2)T,\ \sigma^2 T\right), \qquad \text{median} = S_0 e^{(\mu - \frac12\sigma^2)T},\quad E[S_T] = S_0 e^{\mu T}09Where it fits
Rather not code the maths yourself? Describe your trading idea and let AI build the bot.
Build a bot with AgenKitThe worked example and the model run the same code, so every number matches. Models simplify markets on purpose. This is education, not financial advice.