I'm trying to answer a question about expectations. Let's say I have a random variable, $X$, which is distributed $Expo(\lambda)$. Now assume that $\lambda$ is also an R.V. and is distributed $Expo(5)$. How do I find $E(X)$? Assume that I don't need to do this analytically, I can use the integrate command in R, for instance. I believe I should be using Adam's Law, but I'm not sure. Thank you!

  • 2
    $\begingroup$ This sounds like a self-study question. Can you explain where you have problem with using Adam's Law? $\endgroup$ – Xi'an Feb 12 '18 at 11:34
  • $\begingroup$ I understand the Adam's Law part of it I think (based on Siong's answer below). I'm unsure of how to actually integrate it in R. $\endgroup$ – lish Feb 12 '18 at 16:24
  • $\begingroup$ Why would you want to find a numerical approximation when you can get the exact answer? $\endgroup$ – Xi'an Feb 12 '18 at 17:29

\begin{align} E[X] &= \int_0^\infty E[X|\lambda]f(\lambda) \, d\lambda \\ &= \int_0^\infty E[X|\lambda](5)\exp(-5\lambda) \, d\lambda \end{align}

Hopefully you can take it from here.

Hint: it is something huge.


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.