So suppose $X$ is a rv with a Poisson distribution with $\lambda$ being a random variable as well. $\lambda$ has an exponential distribution with mean $1/c$ and $f_\lambda(t) = c\times\exp(-ct)1_{[0, \infty)}(t)$ where $1_{[0, \infty)}$ is the indicator function of $t$. The question is, what is the distribution of $X$? I really want to simply plug it in, but I feel like that is incorrect. Also, what is the distribution of $\lambda$ has a gamma distribution with scale parameter $\alpha$ s.t. $f_\lambda(t)=c(ct)^{\alpha-1}exp(-ct)/\Gamma(\alpha)$ for $t>0$?

I apologize for the poor formatting, but I couldn't figure out how to format it correctly, and it kept screwing up.


You can think of it like this. What you have initially is a conditional distribution, that is

$$ X|\lambda\sim Po(\lambda) $$

What you want is the marginal distribution of $X$. To this end, remember the following relationship:

$$ p(x)=\int_0^\infty p(x,\lambda)d\lambda=\int_0^\infty p(x|\lambda)p(\lambda)d\lambda $$ where you are given $p(x|\lambda)$ and $p(\lambda)$ in the exercise. So multiply them together and integrate out $\lambda$ and there you go.


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.