Cox process
The mixed Poisson distribution is a discrete probability distribution, which is found as a general approach for the damage index distribution in insurance mathematics. It generalizes the Poisson distribution.
Definition
A random variable satisfies the mixed Poisson distribution with the density when the probabilities
Possesses. If we denote the probabilities of the Poisson distribution, therefore applies
Properties
- The variance is always greater than the expected value. This property is called overdispersion ( engl. Over dispersion). This is in contrast to the Poisson distribution, are the same for the expectation value and variance.
- In practice only be used as densities densities of gamma distributions, log- normal distributions and of the inverse Gaussian distributions.
- Is the density of a gamma distribution, the Poisson distribution is mixed a negative binomial distribution.
In the following, the expected value of the density, and the variance of the density.
Expected value
The expected value is given by
Variance
For the variance we obtain
Standard deviation
From expectation and variance gives the standard deviation
Coefficient of variation
For the coefficient of variation results in:
Skew
The skewness can be represented as
Characteristic function
The characteristic feature is in the form
Generating function
For the generating function is obtained
Moment generating function
The moment generating function of the mixed Poisson distribution is
- Probability distribution