Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I'm revising for an exam and have no idea how to approach this question:

Let $\{N_t\}_{t\geq 0}$ be a homogeneous Poisson process of parameter $\lambda > 0$. Let $\{X_k\}_{k\geq 0}$ be a sequence of random variables, identically independently distributed, with $E[X_k] = 0$ and $Var[X_k] = \sigma^2 < +\infty$. Is the process defined by $Y_t = X_{N_t}$ strictly stationary?

I'm guessing the fact that the Poisson process has independent increments and we have:

$P(N_{t_2} - N_{t_1} = k) = \dfrac{\lambda (t_2 - t_1))^k}{k!} e^{-\lambda (t_2 - t_1)}$ is important, but not sure how it relates, any hints appreciated!

share|improve this question

If the process $(N_t)$ and the sequence $(X_k)$ are independent, then indeed the identity $Y_t=X_{N_t}$ defines a strictly stationary process $(Y_t)$.

To show this, consider $Y^s_t=Y_{s+t}$, then, for every nonnegative time $s$, conditionally on $N_s=m$, the process $(Y^s_t)_{t\geqslant0}$ is such that, for every $t\geqslant0$, $$ Y^s_t=X_{m+N'_{t}},\qquad N'_t=N_{t+s}-N_s. $$ The process $N'$ is independent on $N_s$ hence $Y^s$ is independent on $N_s$. Furthermore, for every $m$, $(X_{m+k})_{k\geqslant0}$ is distributed like $(X_k)_{k\geqslant0}$ hence $(X_{m+N'_t})_{t\geqslant0}$ is distributed like $(X_{N_t})_{t\geqslant0}$ and the result follows.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.