I don't know how should I prove a random variable obey a certain distribution. For instance in the following example, how should I start the proof?

Example: if the number of random points on the axis T which are less than $t_0$ obey poisson distribution with parameter $\lambda t_0$ and the random variable X denote the interval between the first random point on the T axis greater than $t_0$, then X obey exponential distribution.

  • 1
    $\begingroup$ The interval between the first random point.... and what? An 'interval' requires two endpoints... $\endgroup$ – Pierpaolo Vivo Mar 6 '16 at 13:15
  • $\begingroup$ between $t_0$ and the first random point greater than $t_0$ $\endgroup$ – user3070752 Mar 6 '16 at 13:24

I hope I understand well.

Event $X>x$ is the event that the number of random points in interval $(t_0,t_0+x]$ equals $0$.

Probability: $P(N=0)=e^{-\lambda x}$ where $N$ has Poisson distribution with parameter $\lambda$.

So $P(X>x)=e^{-\lambda x}$ making clear that $X$ has exponential distribution with parameter $\lambda$.

  • $\begingroup$ I think I found another solution like your solution. If the random variable N denotes the number of random point in the interval $(0,t_0)$ then probability of occurrence of a random point in a given interval is proportional to the interval length so $P(t<X<t+\delta t)=\lambda \delta t$, then we can use some computation to get $P(X<x)=1-e^{-\lambda x}$ $\endgroup$ – user3070752 Mar 6 '16 at 14:20
  • $\begingroup$ What if $\lambda\delta t>1$? Probabilities do not exceed $1$. $\endgroup$ – drhab Mar 6 '16 at 14:30
  • $\begingroup$ I think the right expression is $P(t<X<t+\delta t) = \lambda \delta t + O(\delta t)$, when $O(\delta t)$ decrease more quickly than $\delta t$. $\endgroup$ – user3070752 Mar 6 '16 at 14:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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