1. Not finding help here? Sign up for a free 30min 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!

Sequences limit law problem

  1. Jan 2, 2014 #1
    Hello everyone,
    I'm starting to study sequences.
    I'm on Stewart's Calculus textbook (single variable, 7th edition, for those who have it, on p. 693).
    Now, I'm at the part where the limit laws are "transferred" to sequences.

    (I'm sorry. I do not know how to code. I hope this is clear to you.)

    lim as n->infinity of anp=[lim as n->infinity of an]p
    IF p>0 AND an>0

    Why do they add the IF p>0 AND an>0? I do not see why the formula would be wrong without that.
     
  2. jcsd
  3. Jan 2, 2014 #2

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Can you think of a counterexample if p = -1?

    What could happen to a_n that would prevent the formula from being true?

    Also, if a_n altenates between +1 and -1, can you see a counterexample?
     
  4. Jan 2, 2014 #3
    Why would a_n alternating change anything?
    I tried to think about what would happen, but everything still makes sense.
    Can you give me the counterexample you are thinking of? I'd like to graph it.
     
  5. Jan 2, 2014 #4

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    [itex]If \ a_n = -1^n[/itex]

    Then what is:

    [itex]lim_{n→∞}a_n[/itex]
     
  6. Jan 2, 2014 #5
    lim of a_n as n->infinity is -1
    lim of (a_n)^(-3) as n->infinity is -1
    (lim of a_n as n->infinity)^(-3) is -1
    The law seems to work! I think I'm doing something clearly wrong.
     
  7. Jan 2, 2014 #6

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    In this case, the limit does not exist. The sequence alternates between 1 and -1 doesn't get stay close to either. So:

    [itex](lim_{n→∞}a_n)^2[/itex] does not exist

    But:

    [itex]lim_{n→∞}a_{n}^2 = 1[/itex]
     
  8. Jan 2, 2014 #7
    Did you mean a_n=(-1)^n? I think if a_n=-1^n, the limit is -1.
     
  9. Jan 2, 2014 #8

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Why would it be -1 in preference to 1? The sequence goes:

    -1, 1, -1, 1, -1, 1, -1, 1...

    Why would that converge to -1?
     
  10. Jan 2, 2014 #9
     
  11. Jan 2, 2014 #10

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    Yes, PeroK meant the former. The expression -1^n is a bit ambiguous as some people will read it as -(1^n) and others will read it as (-1)^n.
     
  12. Jan 2, 2014 #11
    Awesome! Thanks a lot. I'm starting to get this. Can I extrapolate that the a_n being >0 only adresses the issue of signs? What about the fact that p>0? I searched the web and no document explains it. Videos online, including KhanAcademy does not seem to address the issue!
     
  13. Jan 2, 2014 #12

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    For the other one, try a_n = n and p = -1:

    [itex]lim_{n→∞}a_{n}^{-1} = lim_{n→∞}\frac{1}{n} = 0[/itex]

    But:

    [itex]lim_{n→∞}a_{n} = lim_{n→∞}n[/itex] does not exist

    so:

    [itex](lim_{n→∞}n)^{-1}[/itex] is not defined
     
  14. Jan 2, 2014 #13

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    PeroK already gave you a counterexample with p=2 and ##a_n = (-1)^n##. In that case, ##a_n^p = 1##, so the limit on the lefthand side exists. The limit on the righthand side, however, doesn't.

    What about if p=1/2 but ##a_n## wasn't necessarily positive?

    What about if p=-1 and ##a_n \to \infty##?

    What about p=0 and ##a_n=n##?

    Can you see why the equality wouldn't hold in each of these cases?
     
  15. Jan 2, 2014 #14
    Thanks PeroK!
    Vela:
    What about p=0 and a_n=n? This seems to work:
    Left-hand limit=1
    Right-hand limit=1
     
  16. Jan 2, 2014 #15

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    Really? What's ##\lim_{n \to \infty} n##?
     
  17. Jan 2, 2014 #16
    Infinity, no?
     
  18. Jan 2, 2014 #17

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    Better to say the limit doesn't exist.
     
  19. Jan 2, 2014 #18
    True, but doesn't the answer give infinity^0=1?
     
  20. Jan 2, 2014 #19

    vela

    User Avatar
    Staff Emeritus
    Science Advisor
    Homework Helper
    Education Advisor

    No. You can't raise something that doesn't exist to any power.
     
  21. Jan 2, 2014 #20

    PeroK

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    I'm sure the book you have will help with this, but one of the best things you can do before you go much further is:

    1) Understand why ∞ is not a number. And, in particular, that no arithemtic operations are defined on it.

    2) Understand precisely what we mean by n → ∞ and why this does not require ∞ to be a number.

    This is very important before you go very far with real analysis.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted