0
$\begingroup$

Consider a couple of real random variables $(X,Y)$ and $\mu_{X,Y}$ the induced probability distribution. Denote by $\mu_X$ and $\mu_Y$ the distributions of $X$ and $Y$.

Is it true that $$\int_{\mathbb{R}^2}f(x)\ d\mu_{X,Y}(x,y)=\int_{\mathbb{R}}f(x)d\mu_{X}\quad ?$$ This is true each time you can use Fubini theorem. For example, if the random variables are discrete or indepedent or if $(X,Y)$ admits a density with respect to the Lebesgue measure.

I don't know if it is true in the general case, I don't find any counterexample.

Thanks for helping me.

$\endgroup$
2
  • $\begingroup$ isn't by definition $P(x)=\int P(x,y)dy$, so $\int f(x)P(x,y)dxdy=\int f(x)P(x)dx$ $\endgroup$ Nov 5, 2012 at 13:41
  • $\begingroup$ What is your assumption about $f$? $\endgroup$ Nov 5, 2012 at 14:23

0

Your Answer

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

Browse other questions tagged or ask your own question.