How to get the marginal distribution by splitting the integral of one of two variables

73 Views Asked by At

I am trying to follow the answer to this question. Sorry to ask to read another post, but I don't understand why the integration of X is not just between the limits [0,1], where X has support. Instead there is a second integral between 0 and 1/y. This is confusing and the recommendation of drawing the integration area is not helping. I get it that 1/y is meant to find the inverse of U[0,1/x], but in this case, why integrate [0,1] separately?

1

There are 1 best solutions below

0
On BEST ANSWER

Plotting should help see why calculating the marginal from the bivariate needs to be broken down into two different intervals:

enter image description here