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: Properly Divergent Sequences

  1. Oct 29, 2008 #1
    1. The problem statement, all variables and given/known data
    Suppose that (x_n) is a properly divergent sequence, and suppose that (x_n) is unbounded above. Suppose that there exists a sequence (y_n) such that limit (x_n * y_n) exists. Prove that (y_n) ===> 0.

    2. Relevant equations
    (x_n) ===> 0 <====> (1/x_n) ===> 0

    3. The attempt at a solution
    One can say with certanty that (y_n) must be bounded, as if it weren't, for all K in Naturals, there exists a b_1 in (x_n) > |K| and b_2 > |K|, and there product is unbounded.

    If (y_n) is bounded, and does not converge to 0, then... what?

    That's where I'm stuck. How do I finish this?

  2. jcsd
  3. Oct 29, 2008 #2


    User Avatar
    Science Advisor
    Homework Helper

    If y_n does not converge to zero then there is an e>0 such that for all N there is an n>N such that |y_n|>e. If x_n is unbounded, what does this tell about y_n*x_n?
  4. Oct 30, 2008 #3
    When I was trying to prove it directly (for fun), I met some problems.
    What is a *properly* divergent sequence?( I cannot find its definition in books)
    If x_n is defined as follows:
    x_n = 0 when n is even, x_n = n when n is odd
    is it of such kind?
    If so, define y_n as:
    y_n = 1 when n is even, y_n = 0 when n is odd

    does this gives x_n*y_n = 0, as a counter??
    Last edited: Oct 30, 2008
  5. Oct 30, 2008 #4


    User Avatar
    Science Advisor

    The crucial part of the problem is "suppose that (x_n) is unbounded above". Your example does not satisfy that.
  6. Oct 30, 2008 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    A properly divergent series is one such that

    (mathematics) A series whose partial sums become either arbitrarily large or arbitrarily small (algebraically).

    So turning each partial sum into an element of the sequence, I believe a properly divergent sequence is one in which for all M there exists k s.t. j>k => |xj|>M
  7. Oct 30, 2008 #6
    Thanks..I forgot to search the web.... That would make sense. So a direct proof is also not hard.

    BTW, to HallsofIvy, my x_n do satisfy the unboundedness, IMO.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook