0
$\begingroup$

Let $X_1,\ldots,X_5$ and $Y_1,\ldots,Y_8$ be two independent simple random samples such that $X_i\sim\mathcal N(m = 7,\sigma^2 = 50)$ and $Y_i\sim\mathcal N(m = 5,\sigma^2 = 24)$. Find a real number $c$ such that $𝑃(\bar X−\bar Y > c) = 0.3$

Should be an easy problem but I can’t understand if I’m solving it right. How to find this number c, that probability of $\bar X−\bar Y$ higher than c is equal 0.3? Please help

My attempt: Here I found variance of means and their difference: $Var(\bar X) = \delta^2/n = 50/5 = 10$;
$Var(\bar Y) = \delta^2/n = 24/8 = 3$; $\bar X \sim \mathcal N(7, 10)$;
$\bar Y \sim \mathcal N(5, 3)$; $\bar X−\bar Y \sim \mathcal N(2, 13)$; $(\bar X−\bar Y -2)/(√13) \sim \mathcal N(0, 1)$

$\endgroup$
1
  • $\begingroup$ Please add the self-study tag & read its wiki. Then tell us what you understand thus far, what you've tried & where you're stuck. We'll provide hints to help you get unstuck. Please make these changes as just posting your homework & hoping someone will do it for you is grounds for closing. $\endgroup$ Commented Mar 15, 2022 at 15:46

1 Answer 1

1
$\begingroup$

All of your calculations so far are correct, to finish you can notice that for any real number $c$ : $$\begin{align}\mathbb P(\bar X - \bar Y > c) &= \mathbb P(\bar X - \bar Y - 2 > c - 2)\\ &=\mathbb P\left(\frac{\bar X - \bar Y - 2}{\sqrt{13}} > \frac{c-2}{\sqrt{13}}\right)\\ &=1-\Phi\left(\frac{c-2}{\sqrt{13}}\right) \end{align} $$ Where $\Phi$ is the Cumulative Distribution Function of the standard normal distribution.

Therefore $\mathbb P(\bar X - \bar Y > c) = 0.3$ iff $1-\Phi\left(\frac{c-2}{\sqrt{13}}\right) =0.3$. To solve your problem, you thus need to find $c$ such that $$1-\Phi\left(\frac{c-2}{\sqrt{13}}\right) =0.3 $$

$\endgroup$
2
  • $\begingroup$ Thank you so much for your kind help! Do you know any textbook which adequately explains this material? $\endgroup$
    – Aella
    Commented Mar 15, 2022 at 17:21
  • $\begingroup$ Sadly I don't... Maybe you can try Statistical Inference by Casella and Berger ? It builds most of statistical concepts from first principles, and most opinions I heard about it were quite positive. $\endgroup$ Commented Mar 15, 2022 at 18:14

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.