The normal distribution describes the familiar "bell-shaped" probability distribution. The PDF of a normal distribution with mean `mu` and variance `sigma^2` is
`f_X(x; mu,sigma^2) = 1/sqrt(2 pi sigma^2) exp(-1/{2sigma^2} (x-mu)^2). `
It is symmetric about `mu`, and `sigma^2` controls how "spread out" the PDF is.
We usually denote the normal distribution by `N(mu,sigma^2)`.
Standard normal distribution: `N(0,1)`
A key property of the normal distribution: if `X` is `N(mu,sigma^2)`, and `Y = a+bX`, then `Y` is `N(a+b mu, b^2 sigma^2)`.
The Matlab/Octave function randn generates draws from the standard normal distribution.