A càdlàg process $X$ is a stochastic process for which the paths $t\mapsto X_t$ are right-continuous with left limits everywhere, with probability one. The word càdlàg is an acronym from the French for ``continu à droite, limites à gauche''. Such processes are widely used in the theory of noncontinuous stochastic processes. For example, semimartingales are càdlàg, and continuous-time martingales and many types of Markov processes have càdlàg modifications.

Given a càdlàg process $X_t$ with time index $t$ ranging over the nonnegative real numbers, its left limits are often denoted by \begin{equation*} X_{t-}=\lim_{\substack{s\rightarrow t,\\ s<t}}X_s \end{equation*}for every $t>0$ . Also, the jump at time $t$ is written as \begin{equation*} \Delta X_t = X_t-X_{t-}. \end{equation*} Alternative terms used to refer to a càdlàg process are rcll (right-continuous with left limits), R-process and right-process.

Although used less frequently, a process whose paths are almost surely left-continuous with right limits everywhere are known as càglàd, lcrl or L-processes.