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!

A limsup inequality (showing that the root test is stronger than the ratio test)

  1. Oct 3, 2011 #1
    1. The problem statement, all variables and given/known data

    Show that if [itex]a_n > 0[/itex] for all n,

    [itex]\liminf{\frac{a_{n+1}}{a_n}} \leq \liminf{a_n^{1/n}} \leq \limsup{a_n^{1/n}} \leq \limsup{\frac{a_{n+1}}{a_n}}[/itex]


    2. Relevant equations

    [itex]\liminf{a_n^{1/n}} \leq \limsup{a_n^{1/n}}[/itex]

    [itex]\liminf{\frac{a_{n+1}}{a_n}} \leq \limsup{\frac{a_{n+1}}{a_n}}[/itex]
    These are obvious, by properties of sup and inf.

    Which implies that, to complete the proof, I must show that

    [itex]\liminf{\frac{a_{n+1}}{a_n}} \leq \liminf{a_n^{1/n}}[/itex]
    and
    [itex]\limsup{a_n^{1/n}} \leq \limsup{\frac{a_{n+1}}{a_n}}[/itex]

    I'm focusing right now on the limsup inequality.


    3. The attempt at a solution

    I took ln of each term,

    [itex]\ln{\limsup{a_n^{1/n}}} = \limsup{\ln{a_n^{1/n}}} = \limsup{\frac{\ln{a_n}}{n}}[/itex]

    and

    [itex]\ln{\limsup{\frac{a_{n+1}}{a_n}}} = \limsup{\ln{\frac{a_{n+1}}{a_n}}} = \limsup{(\ln{a_{n+1}}-\ln{a_n})}[/itex]

    This implies that I need to show that

    [itex]\limsup{\frac{\ln{a_n}}{n}} \leq \limsup{(\ln{a_{n+1}}-\ln{a_n})}[/itex]


    I've let

    [itex]M_1 := \limsup{\frac{\ln{a_n}}{n}}[/itex]
    [itex]M_2 := \limsup{(\ln{a_{n+1}}-\ln{a_n})}[/itex]

    So now I just have to show that [itex]M_1 \leq M_2[/itex]

    By the definition of limit, for all ε>0 there exists an N1 in the natural numbers such that for all [itex]n > N_1[/itex] we have

    [itex]M_1 - ε < \frac{\ln{a_n}}{n} < M_1 + ε[/itex]


    Similarly, for all ε>0 there exists an N2 such that for all [itex]n > N_1[/itex] we have

    [itex]M_2 - ε < \ln{a_{n+1}}-\ln{a_n} < M_2 + ε[/itex]


    I think I can say that the epsilons are the same, since for each of these statements, I can choose a specific epsilon and I'm guaranteed that an N1, N2 exist to satisfy the inequalities. However, I don't know how to go from here to show [itex]M_1 \leq M_2[/itex]. It might be trivial, I don't know, but I just don't see it. I've been working on this problem for too long, and it's making me cross-eyed. It was a homework problem that I turned in unfinished, and now I'm just trying to finally figure it out for my own sanity.
     
  2. jcsd
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Can you offer guidance or do you also need help?
Draft saved Draft deleted



Similar Discussions: A limsup inequality (showing that the root test is stronger than the ratio test)
  1. Ratio and root tests (Replies: 4)

  2. Ratio and root tests (Replies: 2)

Loading...