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: Limit Question

  1. Oct 9, 2008 #1
    I'm studying for the GRE thats coming up in a week or two and I came across a problem where the answer given in the book does not make sense to me and I was wondering of someone here could explain it to me.


    Lim as n goes to infinity of X_(n+1) / X_n

    Where X_n = n^n / n!


    So I started by simplifying the expression down to:

    Lim as n goes to infinity of (1 + 1/n)^n

    The book informs me and by some proofs online that this tends toward e. However I was hoping someone could explain this to me because from my point of view it should just hit 1.

    Since 1/n -> 0, 1+0 = 1, and 1^n is 1 for any arbitrarly high power.
  2. jcsd
  3. Oct 9, 2008 #2


    User Avatar
    Science Advisor

    Use the binomial theorem on (1 + 1/n)^n and see what happens to each term as n becomes infinite. You will end up with the power series for e.
  4. Oct 10, 2008 #3
    So, using binomial expansion we get:

    Lim_n^Inf of
    Series m=0 to n

    (1)^(n-m) * (1/n)^m * nCm

    Note: nCm = n!/[m!(n-m)!]

    I'm a little unsure how to simplify this. Obviously from your response I'm looking to simplify this to 1/m!.

    Assuming I can pass the limit through the series (unsure of this) we get:

    Series m=0 to Infinity
    Lim_n^Inf of
    (1/n)^m * n!/[m!(n-m)!]

    What next?
  5. Oct 13, 2008 #4
    Bump* Just wanted to see if anyone could clarify this question for me.
  6. Oct 14, 2008 #5


    Staff: Mentor

    I thought it would be worthwhile to jump in here and point out that your analysis is faulty in the line just above. Yes, it's true that 1/n --> 0 as n gets large, and 1^n is 1 for any arbitrarily large finite power, but this is not the way limits work. The limit process applies to the whole expression, not just a bit here and another bit there later on.

    The point is that, although 1/n --> 0 in the limit, for any finite value n, 1/n is not zero, so for the same value of n, (1 + 1/n) ^ n is not 1^n.

    So while the base is getting closer to 1, the exponent is getting larger and larger. This type of limit is one of several that are called indeterminate forms. I'm reasonably sure you can do a search on wikipedia to find a page with more information. They are called indeterminate because you can't determine at a glance what their limits will be.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook