Hyper-Erlang distribution

In probability theory, a hyper-Erlang distribution is a continuous probability distribution which takes a particular Erlang distribution E_i with probability p_i. A hyper-Erlang distributed random variable X has a probability density function given by

${\displaystyle A(x)=\sum _{i=1}^{n}p_{i}E_{l_{i}}(x)}$

Diagram showing queueing system equivalent of a hyper-Erlang distribution

where each p_i > 0 with the p_i summing to 1 and each of the E_li being an Erlang distribution with l_i stages each of which has parameter λ_i.^[1][2][3]

See also

• Phase-type distribution