4
$\begingroup$

I'm taking an upper level Economics class and one of my assignments asks the question in the title.

I approached it by using one property of expectation: expectation of the sum is equal to expectation of it's parts.

So, I did:

$$E[(2X + 3)^2] = E[4X^2 + 12X + 9] = 4E[X^2] + 12 E[X] + E[9]$$

I didn't get the right answer. I'm not even sure if what I did was the approach.

There's also second part to this question, which also asks for the variance VAR(2X + 5), given VAR(X) = 4. I can only think of one relevant property of the variance and it doesn't help with that question.

Anyone have an idea of how I should approach this?

$\endgroup$
  • 3
    $\begingroup$ You will need information about $E[X^2]$. Is there anything in the original problem statement that can help you figure out that value? $\endgroup$ – whuber Apr 14 '15 at 16:01
  • $\begingroup$ Ha. @whuber corrects the maths while latexing it. $\endgroup$ – conjugateprior Apr 14 '15 at 16:02
  • $\begingroup$ @conjugate I'm not trying to sneak in an answer via an edit: I merely corrected some typographical errors in an intermediate expression. The distinction is important, because a true mathematical error in a question usually gives important clues concerning why the question arose and how to answer it. Therefore, in editing a question, we should usually take care not to fix actual errors. $\endgroup$ – whuber Apr 14 '15 at 16:06
  • 2
    $\begingroup$ Is $X$ the same variable in part 1 and 2? $\endgroup$ – ekvall Apr 14 '15 at 18:02
  • 1
    $\begingroup$ "I originally thought that E[X^2} for this problem would also be 1" No, except in degenerate cases (when $X$ is a constant), $E[X^2] \gt (E[X])^2$. The difference $E[X^2] - (E[X])^2$ is the variance of $X$. But to repeat whuber's question with slightly different wording "Are you sure you are not told anything else about $X$"? For example, you might not be sharing with us that the information that $X$ is an exponential random variable on the grounds that it is irrelevant, but it really is a very relevant fact. For an exponential random variable with $E[X]=1$, $E[X^2]$ equals $2$. $\endgroup$ – Dilip Sarwate Apr 14 '15 at 18:55
4
$\begingroup$

Based on the comments, I assume the that the $X$ in part 1 and 2 is the same. Then we know $\mathbb E X =1,\mathrm{Var}(X)=4$.

You may now find the answer by using the relationship $\mathrm{Var}(X)=\mathbb EX^2- (\mathbb E X)^2$. (Hint: The correct answer is 41.)

I leave the below as an example of why the information in the first part is not sufficient.

Let $Y=1$, $X\sim N(1,1)$. Then $\mathbb E (2Y+3)^2=25$ while $(2X+3)=:Z\sim N(5, 4)$ so that $\mathbb E Z^2 = \mathrm{Var}(Z)+(\mathbb E Z)^2=29$.

$\endgroup$
  • $\begingroup$ X looks to be the same variable in part 1/2, and before I read your comment I ended up getting 16 as an answer after realizing that Var (aX+B) = a^2 Var(X). I think that's the same as Var(X)= E[X]^2- (EX)^2? Would I also be wrong in getting 25 for the answer in the first part? The professor might have made a typo in the answer choices he gave us, which has happened a lot before. $\endgroup$ – user1943854 Apr 14 '15 at 18:53
  • 1
    $\begingroup$ Re: "Var (aX+B) = a^2 Var(X). I think that's the same as Var(X)= E[X]^2- (EX)^2?" No, they're not the same formula. 25 is not the correct answer for the first question. Have you tried using the formula I gave for the variance together with the given fact $Var(X)=4$ on the expression you have derived in your question? $\endgroup$ – ekvall Apr 14 '15 at 19:04
  • 1
    $\begingroup$ You're right, I misread the notes and the formulas aren't the same. Yes, I thought it over some more and from what you gave me, I realized that E[X]^2 = 5. Would that be wrong? If I plug that into part 1 and I get 41, which is D from the answer choices. $\endgroup$ – user1943854 Apr 14 '15 at 19:13
  • 1
    $\begingroup$ You are correct! $\endgroup$ – ekvall Apr 14 '15 at 19:15
  • 1
    $\begingroup$ Your formula is correct for part 2 $\endgroup$ – ekvall Apr 14 '15 at 19:23

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.