2

Can one use convolutions to construct the density of a sum of dependent variables, and if so, how?

I understand that to construct the density of the sum of two possibly dependent random variables, it is possible to apply the convolution formula using their joint density (e.g., How to add two dependent random variables?).

If one needs to construct the pdf for the sum of multiple independent variables, it is possible to take convolutions iteratively. But I wouldn't know how to proceed for multiple dependent variables.

Community
  • 1
cheby
  • 21
  • 1
  • Welcome to our site! I have made some edited the title to make it slightly more informative - feel free to edit to improve further, or roll my edit back if you dislike it. (There's no need to add a "thanks in advance" at the end - we try to cut out the pleasantries from the questions, just so that future readers have less to read through. But you can always thank someone who answers your question by upvoting them and giving the "green tick" of acceptance to the answer you find most helpful.) – Silverfish Aug 01 '16 at 22:41
  • 1
    The short answer is you need to integrate joint densities. Can you specify the distributions of your dependent variables? – Alex R. Aug 01 '16 at 23:33

0 Answers0