The Claude Quant PlaybookDownload ebookDownload ebook

Normal distribution

beginnerDistributions7 of 11

The bell curve. It is fixed by two numbers, the mean and the standard deviation, and the share of outcomes within k standard deviations of the mean is the same for every normal.

Build a trading system with Claude.Download ebook

01The formula

the density: height of the bell at x
the mean, the centre of the bell
the standard deviation, the width of the bell
how many standard deviations either side of the mean
the standard normal CDF: the chance a standard bell-curve draw is below k

02See it move

03Where it is used

04Why traders use it

Averages of many small independent shocks are close to normal, so it is the natural first model for returns. Knowing its 68–95–99.7 rule lets you size a move in seconds.

05Worked example

Daily returns with mean 0.05% and volatility 1.0%: 95.45% of days fall within 2.00 standard deviations, between −1.95% and 2.05%. The other 4.55% of days are outside.

06How it is built

  1. Standardise: is a standard normal with mean 0 and standard deviation 1.
  2. So is the same event as .
  3. Its probability is .
  4. By symmetry , giving : 68.3% for , 95.4% for , 99.7% for .

07The common mistake

Trusting the normal in the tails. Real daily returns have far more 4σ and 5σ days than the bell curve allows, so risk built only on it understates crashes.

08In Python and LaTeX

from scipy.stats import norm
mu, sigma, k = 0.0005, 0.01, 2
print(round(norm.cdf(k) - norm.cdf(-k), 4), mu - k*sigma, mu + k*sigma)
f(x) = \frac{1}{\sigma\sqrt{2\pi}}\, e^{-\frac{(x - \mu)^2}{2\sigma^2}}, \qquad P(|X - \mu| < k\sigma) = 2N(k) - 1

09Where it fits

Rather not code the maths yourself? Describe your trading idea and let AI build the bot.

Build a bot with AgenKit

The worked example and the model run the same code, so every number matches. Models simplify markets on purpose. This is education, not financial advice.