A general method for the simulation of observations of a continuous random variable X.
Let u be a pseudo-random number and let F be the distribution function of X. Then x, given by
is a random observation of X. For example, an exponential random variable X, with mean μ, has distribution function (1 − e−x/μ); hence − μ ln (1−u) is an observation of X.
Subjects: Probability and Statistics.