I have trouble with conditional probabilities, therefore I am wondering if the following derivation is correct:

Both X and Y are exponential random variables, with $\lambda_x$ and $\lambda_y$ respectively as their rates, and are independent of each other. Then $E[X | X > Y]$ = $ \int_0^\infty X f_{X|X>Y}(x)dx$ = $ \int_0^\infty X (1-f_{X|X<Y}(x))dx$ = $ \int_0^\infty Xdx$ - $ \int_0^\infty X f_{X|X<Y}(x)dx$ = $ 1 - \frac{1}{\lambda_x +\lambda_y} $.

I remember the property $E[X | X < Y]$ = $E[min(X,Y)]$ = $\frac{1}{\lambda_x +\lambda_y} $. But it just seems wrong to me, that you get 1 - $E[X | X < Y]$ as answer for $E[X | X > Y]$, as if it is a probability. So I assuming that I am making an error.

  • $\begingroup$ How did you arrive at "$\int_0^\infty X dx = 1$"? If you really mean capital $X,$ then this expression makes no sense; and if you mean lower case $x,$ then it diverges. Before you worry about this too much, could you explain why you believe $f_{X\mid X\gt Y} = 1- f_{X\mid X\lt Y}$? How would you interpret this for values of $f$ that exceed $1$ (which will happen when rates exceed $1$)? $\endgroup$ – whuber Feb 7 at 16:02

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.