Poisson

Overview


The poisson distribution measures the probability of a number of events ocurring. As such, it is a probability over a discrete non-negative variable.

Definition


The probability distribution is given by the following
{% P(x) = \lambda^x e^{-\lambda} / x! %}
where {% x %} is an non-negative integer, that is
{% x %}=0,1,2,3.....
This is sometimes re-written in exponential form as
{% P(x) = exp[x log(\lambda) - \lambda - log \Gamma(x+1)] %}
Note, some authors use {% \mu %} in place of {% \lambda %} in the formulas above.

Moments


The moments of the distribution are given by
{% \mathbb{E}[x] = \lambda %}

{% \sigma^2 = \lambda %}

Probability of No Events


The probability that no events have ocurred up to time {% t %} is given by
{% P = exp(- \lambda t) %}

Sum of Poisson Variables


Given two Poisson distributions, {% L_1 \sim Pois(\lambda_1) %} and {% L_2 \sim Pois(\lambda_2) %} we have
{% L_1 + L_2 \sim Pois(\lambda_1 + \lambda_2) %}

Poisson Library


The poisson library provides functionality for doing computations with the poisson distribution.


	/lib/statistics/distributions/poisson/v1.0.0/poisson.mjs