## Glossary

A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

# Compound Poisson process

A compound Poisson process is a Poisson process where we let the jumps be of random height instead of always one as in the ordinary Poisson process. A compound Poisson process is an important special case of the more geberal class of Lévy processes. A compound Poisson process can also be reprsented using a discrete time random walk a Poisson process. Let $\{X(k)\}_{k=0,1,\ldots}$ be a discrete time random walk and let $N$ be a Poisson process. We can now obtain a compound Poisson process $Z$ by setting $Z(t)=X(N(t))$. The characteristic function of compound Poisson proces $X(t)$ is $\phi_{X(t)}=E[e^{iyX(t)}]=e^{tK(y)},$ where $K(y)=iy\mu+\lambda(\phi_J(y)-1)$, where $\phi_J$ is the characteristic function for the jump distribution which we can choose freely.