Find E[(X-mu)^k] - Normal Random Variable

  • Thread starter motherh
  • Start date
  • #1
27
0

Main Question or Discussion Point

Hi,

I'm having a bit of a problem with a probability question. The question is

Let X be a normal random variable with mean [itex]\mu[/itex] and variance [itex]\sigma^{2}[/itex]. Find E[(X -[itex]\mu[/itex])[itex]^{k}[/itex]] for all k = 1,2,...

I'm not really sure what to do and need some help to confirm how to approach the question. I've tried messing around with moment generating functions, using the binomial theorem to expand E[(X -[itex]\mu[/itex])[itex]^{k}[/itex]] and also have tried computing the expectation directly. Are any of these methods the correct way to progress?

Any help is much appreciated, thanks!
 

Answers and Replies

  • #2
269
24
Letting [itex]m_k= \mathbb E[(X-\mu)^k][/itex], we know [itex]m_1=0[/itex] by definition of [itex]\mu[/itex]. Next, maybe try to use integration by parts (and the exact functional form of the p.d.f. [itex]\phi[/itex]) to express [itex]m_k[/itex] ([itex]k=2,3,...[/itex]) in terms of [itex]m_1,...,m_{k-1}[/itex].
 
  • #3
27
0
Hi, thanks for your help so far! I'm a little bit confused though. What do I integrate by parts exactly? And I'm a little bit thrown by "the exact functional form of the p.d.f. ϕ" - what do you mean? Apologies, I use ϕ as part of my notation for moment generating functions.
 
  • #4
27
0
Anyone? I'm still really confused. Do I need a binomial expansion of (X-mu)^k in my answer somewhere?
 
  • #5
269
24
Hey. I was using the notation [itex]\phi(z) = \frac1{\sqrt{2\pi}}e^{-\frac{z^2}2}[/itex], the probability density function of [itex]\mathcal N(0,1)[/itex], so that the the p.d.f. of a [itex]\mathcal N(\mu,\sigma^2)[/itex] random variable is [itex]f(x)=\phi\left(\frac{x-\mu}{\sigma}\right)[/itex].

Then, we can express the desired numbers as [tex]m_k = \int_{-\infty}^\infty (x-\mu)^k f(x)\text{ d}x.[/tex]
 
  • #6
269
24
To clean up notation, let's notice that [itex]m_k = \sigma^k n_k[/itex], where [itex]n_k = \mathbb E \left[ \left( \frac{x-\mu}{\sigma} \right)^k \right][/itex]. So we can just work with the system [tex]n_k = \int_{-\infty}^\infty z^k\phi(z)\text{ d}z,[/tex] which I would suggest attacking with IBP.
 
  • #7
27
0
Ah, thank you so much! I got the relation m_(k+2) = sigma^2 * (k+1) * m_k, does that sound right?

With m_1 = 0 it means m_k = 0 for odd k which does seem to make sense.
 

Related Threads on Find E[(X-mu)^k] - Normal Random Variable

Replies
1
Views
2K
  • Last Post
Replies
6
Views
22K
  • Last Post
Replies
14
Views
15K
Replies
10
Views
950
  • Last Post
Replies
5
Views
2K
Replies
9
Views
11K
Replies
1
Views
2K
Replies
7
Views
3K
Top