1: \begin{definition}

2: A {\em L\'evy process} $L$ in $\R^d$ is a stochastic process

3: in continuous time, defined on some probability space $(\Omega,\cF,\P)$, 

4: having the following properties:

5: 

6: \begin{enumerate}

7: \item $L_0=0$ almost surely;

8: \item $L$ has {\em stationary increments}, that is, for any $s,t\ge 0$,

9: the random variables $L_{s+t}-L_t$ and $L_s$ have the same distribution; 

10: \item $L$ has {\em independent increments}, that is, 

11: for any $n\in\N$ and any $t_1,\ldots,t_n\in [0,\infty)$ with 

12: $t_0:=0< t_1<t_2<\ldots<t_n$, 

13: the random variables $L_{t_1}-L_{t_0},\ldots,L_{t_n}-L_{t_{n-1}}$ are

14: independent;

15: \item $L$ is {\em continuous in probability}, i.e., for all $t\ge 0$ and

16: $c>0$,

17: \[

18: \lim_{h\rightarrow 0}\P(|L_{t+h}-L_t|>c)=0\,.

19: \]

20: \end{enumerate}

21: Without loss of generality one may also require (see \cite[Chapter I.4, Theorem 30]{protter})

22: \begin{enumerate}

23: \setcounter{enumi}{4}

24: \item $L$ has {\em c\`adl\`ag paths}, that is, for each $\omega\in \Omega$ 

25: the mapping 

26: $t\longmapsto L_t(\omega)$ is right-continuous with limits from the left.

27: \end{enumerate}

28: \end{definition}

29: