-1
$\begingroup$

This answer says that if $X$ is a random variable and $X_+ = \mathrm{max}(0, X)$, then $X_+ = \int_0^\infty I_{\{X > x\}}\mathrm{d}x$. I'd like to know how to derive this starting with $A \in \mathcal{S} \implies \int_S 1_A\mathrm{d}\mu = \mu(A)$ (from "Desired Properties").

My thinking is that for a probability space $(\Omega, F, \mu)$, $\{X > t\} \in F \implies \int_\Omega I_{\{X > t\}}\mathrm{d}\mu = \mu(\{X > t\})$ but it's the wrong variable of integration. Am I on the right track?

$\endgroup$
3
  • $\begingroup$ Does max(0, X) mean max{0, X}, i.e., the larger of X and 0 ? $\endgroup$ Mar 25, 2023 at 1:01
  • $\begingroup$ Equality $X_+ = \int_0^\infty I_{\{X > x\}}\mathrm{d}x$ is obvious and is an equality between random variables. It cannot follow from equalities between expectations. Moreover, Mathoverflow is devoted to questions about research. $\endgroup$ Mar 25, 2023 at 9:22
  • 1
    $\begingroup$ Just to amplify what Christophe said: $y_+=\int_0^\infty I_{\{y>x\}}\mathrm{d}x$ is simply true for any real number $y$. There is nothing special involving random variables going on. $\endgroup$ Mar 25, 2023 at 11:43

1 Answer 1

0
$\begingroup$

The layer cake representation of a non-negative measurable function, $X$, is applied in the proof of proposition 2.1 here.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.