1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: Computing the sum of a particular series.

  1. Jan 29, 2013 #1
    1. The problem statement, all variables and given/known data

    3. The attempt at a solution

    Alright, so, I'm clueless about doing this one. I do know that it's extremely similar to e^x


    But really, that means nothing! Usually there's another function/series I can compare and then integrate, in this case I don't have any idea on how to procceed.

    Can someone tip me in the right direction, please?

    Thanks in advance!
  2. jcsd
  3. Jan 29, 2013 #2


    User Avatar
    2017 Award

    Staff: Mentor

    Those quotation marks are "n"?

    ##\sum_{n=0}^\infty \frac{(x-1)^n}{(n+2)!} = \sum_{n=2}^\infty \frac{(x-1)^{n-2}}{n!}
    = \frac{1}{(x-1)^2} \sum_{n=2}^\infty \frac{(x-1)^n}{n!}##
    Just 2 steps left to get an exponential function.
  4. Jan 29, 2013 #3
    You can try to use the Poisson distribution find the solution from there.

    In other words,

    multiply your summation with the e^(-x)/(e^(-x)). Move the denominator out. We know that the summation(e^(-x) * x^n/ n!) must be 1 since it is a distribution (Poisson). So the answer is 1/(e^-x).
  5. Jan 29, 2013 #4
    I think they're "n", I'm not absolutely sure, but since the ebook was scanned with auto recognition, it's very possible that when scanned the "n" was with a poor print, and it recognized as ", it's the only explanation I could come up with.

    Regarding the answer, whoa, I never heard of shifting the start of the sum to do that, it was very clever, so the answer would be:
    \frac{1}{(x-1)^2} \sum_{n=2}^\infty \frac{(x-1)^n}{n!} = \frac{1}{(x-1)^2}*(e^{(x-1)} - \frac{x-1}{6} - \frac{1}{2}) = \frac{-ex+6e^x-2e}{6e(x-1)^2}##

    Is this correct?

    I haven't learned Poisson distribution yet, I'm going to see if I find this further in the book.
  6. Jan 30, 2013 #5


    User Avatar
    2017 Award

    Staff: Mentor

    I think the denominators for (x-1) and 1 are wrong.
    And I would not split that factor of e at e^(x-1), that looks weird (at least to me).
  7. Jan 30, 2013 #6
    Ohh, I think I get it, I used the wrong term: (N+2)! , I should have used N!

    So in fact:
    \frac{1}{(x-1)^2} \sum_{n=2}^\infty \frac{(x-1)^n}{n!} = \frac{1}{(x-1)^2}*(e^{(x-1)} - \frac{x-1}{1!} - \frac{1}{0!}) = \frac{e^x - ex}{e(x-1)^2}##


    Why do you say it looks weird? You have a more elegant solution in mind?
  8. Jan 30, 2013 #7


    User Avatar
    2017 Award

    Staff: Mentor

    Well, that is just a matter of taste (and you have less e hanging around now), but I would write it as $$\frac{e^{x-1}-x}{(x-1)^2}$$
  9. Jan 30, 2013 #8
    Hmm indeed.

    Well, thanks you for all your help so far mfb! That was great.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook