In probability theory, a compound Poisson distribution is the probability distribution of the sum of a number of independent identically-distributed random variables, where the number of terms to be added is itself a Poisson-distributed variable. In the simplest cases, the result can be either a continuous or a discrete distribution.
Contents
- Definition
- Properties
- Discrete compound Poisson distribution
- Other special cases
- Compound Poisson processes
- Applications
- References
Definition
Suppose that
i.e., N is a random variable whose distribution is a Poisson distribution with expected value λ, and that
are identically distributed random variables that are mutually independent and also independent of N. Then the probability distribution of the sum of
has a well-defined distribution. In the case N = 0, then the value of Y is 0, so that then Y | N = 0 has a degenerate distribution.
The compound Poisson distribution is obtained by marginalising the joint distribution of (Y,N) over N, where this joint distribution is obtained by combining the conditional distribution Y | N with the marginal distribution of N.
Properties
Mean and variance of the compound distribution derive in a simple way from law of total expectation and the law of total variance. Thus
giving
Then, since E(N)=Var(N) if N is Poisson, and dropping the unnecessary subscripts, these formulae can be reduced to
The probability distribution of Y can be determined in terms of characteristic functions:
and hence, using the probability-generating function of the Poisson distribution, we have
An alternative approach is via cumulant generating functions:
Via the law of total cumulance it can be shown that, if the mean of the Poisson distribution λ = 1, the cumulants of Y are the same as the moments of X1.
It can be shown that every infinitely divisible probability distribution is a limit of compound Poisson distributions. And compound Poisson distributions is infinitely divisible by the definition.
Discrete compound Poisson distribution
When
has a discrete compound Poisson(DCP) distribution with parameters
Moreover, if
Feller's characterization of the compound Poisson distribution states that a non-negative integer valued r.v.
This distribution can model batch arrivals (such as in a bulk queue ). The discrete compound Poisson distribution is also widely used in actuarial science for modelling the distribution of the total claim amount.
When some
has a discrete pseudo compound Poisson distribution with parameters
Other special cases
If the distribution of X is either an exponential distribution or a gamma distribution, then the conditional distributions of Y | N are gamma distributions in which the shape parameters are proportional to N. This shows that the formulation of the "compound Poisson distribution" outlined above is essentially the same as the more general class of compound probability distributions. However, the properties outlined above do depend on its formulation as the sum of a Poisson-distributed number of random variables. The distribution of Y in the case of the compound Poisson distribution with exponentially-distributed summands can be written in an form.
Compound Poisson processes
A compound Poisson process with rate
where the sum is by convention equal to zero as long as N(t)=0. Here,
For the discrete version of compound Poisson process, it can be used in survival analysis for the frailty models.
Applications
A compound Poisson distribution, in which the summands have an exponential distribution, was used by Revfeim to model the distribution of the total rainfall in a day, where each day contains a Poisson-distributed number of events each of which provides an amount of rainfall which has an exponential distribution. Thompson applied the same model to monthly total rainfalls.