# Probability of $\lambda$ given k, in Poisson distribution.

We know the PMF of poisson distribution is $P_K(k) = e^{-\lambda} \frac{\lambda ^k}{k!}$, now, given $k$ arrivals in a unit time, what is the PDF of the arriving rate being $\lambda$?

-
I don't know this notation: What does $K$ stands for? –  math Dec 4 '12 at 13:48
It means $P(K=k)$, $K$ stands for the random variable, $k$ stands for the value of $K$. –  qed Dec 4 '12 at 14:07

@CravingSpirit: OK; to put it in more mundane terms: What you think about $\lambda$ after you learn that there were $k$ arrivals in a unit time depends on what you that about $\lambda$ before you learned that. For instance, if you think a priori that very low values of $\lambda$ are very unlikely, then even after you observe $0$ arrivals the probability you assign a posteriori to low values of $\lambda$ will be less than the probability you would have assigned a posteriori after observing $0$ arrivals if you had a priori believed these low values to be quite likely. –  joriki Dec 4 '12 at 14:30
Ok, you mean the probability of $\lambda$ depends both on our a priori judgment and the value of $k$. –  qed Dec 4 '12 at 14:35