We have a question regarding two random variables $X$,$Y$ on a probability space with sigma-algebra $\mathcal{F}$ and a sub-sigma algebra $\mathcal{M}$ such that $X$ is independent of $\mathcal{M}$ and $Y$ is $\mathcal{M}$-measurable.

We are asked to show that $E(I(X+Y \text{ in } A)|\mathcal{M}) = E(I(X+Y \text{ in } A)|Y)$ a.s. where $I$ is the indicator function and $A$ is a Borel set on $\mathbb{R}$.

My attempt: Write $\mathcal{H} = \sigma(Y)$, the sigma algebra generated by $Y$. We have that $\mathcal{H}$ is a subset of $\mathcal{M}$, as $Y$ is $\mathcal{M}$-measurable.

Write $Z=I(X+Y \text{ in } A)$

Now, I want to show that $E[Z|\mathcal{H}]$ is a conditional expectation on $\mathcal{M}$, because then by the almost sure uniqueness of conditional expectation the result follows, and so we have to show firstly that it is a $\mathcal{M}$-measurable function, which follows as it is a $\mathcal{H}$-measurable function, and secondly that for any $A$ in $\mathcal{M}$, $E[E[Z|\mathcal{H}]I(\mathcal{M})]=E[Z I(\mathcal{M})]$.

I can't for the life of me seem to figure out this step, any hints would be appreciated.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.