1
$\begingroup$

I just met this in probability and it got me completely stumped:

We define an i.i.d sequence of normally distributed random variables $ \{ X_n \}_{n=1}^{\infty} $ such that $ X_n \sim \mathcal{N} (0,\sigma ^2) $ and we define a random variable $ N \sim Pois(\Lambda) $ independent of the $ X_n $'s

We are asked to show that $ X_N $ and N are independent random variables

I cannot seem to get past the simple fact that these seem clearly dependent as N clearly appears as a subscript for the X random variable so I do not really know what is wrong or how to prove that these are independent random variables so this is where I need help. Thanks all.

$\endgroup$
2
$\begingroup$

Independence follows from the fact that for every fixed $k\in\mathbb N$:$$P(X_N\in A)=\sum_{n=0}^{\infty} P(X_N\in A\mid N=n)P(N=n)=$$$$\sum_{n=0}^{\infty} P(X_n\in A)P(N=n)=\sum_{n=0}^{\infty} P(X_k\in A)P(N=n)=$$$$P(X_k\in A)=P(X_N\in A\mid N=k)$$

$\endgroup$
-2
$\begingroup$

This is not different from rolling a green and a red die ans using a coin flip to decide which of the dice you use and which you ignore. You still have a coin flip and a die.


All that matters is that the $X_n$ are identically distributed (they need not be normally distributed and not independent of one anaother) and that $N$ is independent of all $(X_1,X_2,\ldots)$.

For any measurable $A$ and $n\in\Bbb N$, we have (letting $Y=X_N$ for clarity) $$\begin{align} P(Y\in A\mid N=n)&=P(X_n\in A\mid N=n)&\text{def of $Y$}\\&=\frac{P(X_n\in A\wedge N=n)}{P(N=n)}&\text{def of conditional prob.}\\&=P(X_n\in A)&\text{independent}\\&=P(X_1\in A)&\text{identically distributed}\end{align}$$ so that (as one might have guessed) $$ \begin{align}P(Y\in A)&=\sum_n P(Y\in A\mid N=n)P(N=n)\\&=\sum_nP(X_1\in A)P(N=n)\\&=P(X_1\in A)\sum_nP(N=n)\\&=P(X_1\in A)\end{align}$$ and hence $$P(Y\in A\mid N=n)=P(Y\in A).$$

$\endgroup$
  • 3
    $\begingroup$ Something seems wrong here. First computations a conditional probability is equal to a product of multiplications maybe you mean AND instead of conditional probability? $\endgroup$ – kroner Apr 19 '16 at 8:36
  • $\begingroup$ @Hagen Something seems wrong with your answer $\endgroup$ – Don John Prep Apr 19 '16 at 8:46
  • $\begingroup$ Your first sentence makes things clear on a nice way. If you repair (e.g. by deleting everything from: "for any measurable $A$...") then I will turn my downvote into an upvote. $P(Y\in A| N=n)$ and $P(Y\in A\wedge N=n)$ are essentially different. $\endgroup$ – drhab Apr 20 '16 at 14:04
  • $\begingroup$ I decided to do it myself. My downvote stays a downvote. $\endgroup$ – drhab Apr 21 '16 at 8:35

Your Answer

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

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