How to find expectation of a composition of PDF?

275 Views Asked by At

I have a probability question. I am struggling with a problem about finding the expectation of a compositions of PDF's. I know that expectation of a PDF is $$E(X)=\sum x p(x, \lambda)$$ I am not sure how to handle if the parameter itself is a PDF too. Here's the question: Give Poisson distribution $p(x, \lambda)=e^{-\lambda} \frac{\lambda^x}{x!}$ with parameter $\lambda$... but $\lambda$ is also a random variable with PDF of $f(\lambda) = e^{-\lambda}$... Find the Expectation.

I have a couple of ideas: 1) just make the "mother pdf" written in 2 variables. 2) somehow find expectation of Y, then plug that into "mother pdf" and find its expectation. I've tried both, but don't seem to be getting anywhere.

Any suggestions would be helpful.

1

There are 1 best solutions below

1
On

Given $\lambda=\ell$, if $ e^{-\ell} \frac{\ell^x}{x!}$ can be considered as the pmf of the corresponding conditional distribution, then the conditional expectation is $\ell$. So, the expectation is $$\int_0^{\infty}\ell e^{-\ell}\ d\ell=1.$$