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!

Cauchy sequnce and convergence of a non-monotonic sequence.

  1. May 9, 2012 #1
    1. The problem statement, all variables and given/known data
    Hello,
    I have a question concerning convergence of the non-monotonic sequences which takes place when the Cauchy criterion is satisfied.
    I understand that [tex]|a_n - a_m| <ε[/tex] for all n,m[tex]N\ni[/tex]

    2. Relevant equations
    What I don't see is how [tex](a_{n+1} - a_n) →0[/tex]is not equivalent/enough. Doesn't the fact that the difference tends to zero mean that it can eventually be smaller than any ε?

    3. The attempt at a solution
    I know the examples when this doesn't work, like [tex](\sqrt{n})[/tex] but I don't understand the difference between these two criteria.

    Could anyone explain why they aren't equivalent?
     
  2. jcsd
  3. May 10, 2012 #2
    So let's take the example of an=√n. For large n, you have then [itex]a_{n+1}-a_n \sim n^{-1/2} \rightarrow 0[/itex] when n→∞. We can write the Cauchy criterion as (let p = n-m)
    [tex] \epsilon > a_{m+p}-a_m = \sum_{n=m}^{m+p} a_{n+1}-a_n [/tex]
    So now when n is very large, we have
    [tex] \epsilon > \sum_{n=m}^{m+p} a_{n+1}-a_n \sim \sum_{n=m}^p n^{-1/2} \sim \int_m^{m+p} dn n^{-1/2} = (\sqrt{m+p}-\sqrt{m})/2 [/tex]
    and this grows without limit if you let p to grow, so it can't be smaller than ε for all p.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Cauchy sequnce and convergence of a non-monotonic sequence.
Loading...