Let $X$ and $Y$ be two independent random variables, where $X\sim \operatorname{Pois}(\lambda_1)$ and $Y\sim \operatorname{Pois}(\lambda_2)$.

It's fairly straightforward to show mathematically that $$P(X=x\mid X+Y=n) = \binom n x \left(\frac{\lambda_1}{\lambda_1+\lambda_2}\right)^x \left(\frac{\lambda_2}{\lambda_1+\lambda_2}\right)^{n-x},$$ which shows that given $X+Y=n$, $X\sim \operatorname{Binom}\left(\frac{\lambda_1}{\lambda_1+\lambda_2},n\right)$.

Is there any intuitive way to understand this result in terms of Poisson point processes, or is this simply a convenient mathematical property?


Suppose this is a Poisson process of rate $1$, with $X$ being the number of events in time $(0,\lambda_1]$ and $Y$ being the number of events in time $(\lambda_1,\lambda_1+\lambda_2]$

Then given the total number of events in the time interval $(0,\lambda_1+\lambda_2]$, since this is a Poisson process then each event can be considered to be in a sense uniformly distributed across that time interval independently of the others and has probability $\frac{\lambda_1}{\lambda_1+\lambda_2}$ of being before time $\lambda_1$ and so the conditional distribution of the number in the first part is binomial

  • $\begingroup$ Dear Professor Mr. Henry , I'm so sorry for this comment..I asked the question mathoverflow for help. But the question was downvoted. I had to delete the question.Now, if I ask in MSE , I know that, the question again will be downvoted. I really need help. But unfortunately, no one helped. If I ask you for help, can you take a look? If you have a few minutes. Then, I will delete this comment..Thank you very much.. mathoverflow.net/q/298997/123863 $\endgroup$ – Mathematics is Life Apr 30 '18 at 11:09

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.