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!

How do these statements differ?

  1. Dec 15, 2013 #1
    More than once I have been marked wrong for this, and it's beginning to annoy me...

    How is the statement,

    [itex] |x| < 1 [/itex]

    different from the statement,

    [itex] |x| \leq s < 1 [/itex] for some [itex] s [/itex] ?

    This comes up when talking about the radius of convergence for some series.
  2. jcsd
  3. Dec 15, 2013 #2
    One statement is a partial of the other.
  4. Dec 15, 2013 #3
    If that's supposed to be a joke or something, I really don't get it. I'm asking how they logically differ and what difference it makes about the statement of convergence.
  5. Dec 15, 2013 #4
    What is the relation to s in the first and only the first statement?
    It does not exist.
    Therefore it is a partial statement of the second one. It usually will be marked wrong.
  6. Dec 15, 2013 #5


    User Avatar
    Science Advisor

    Assuming we are talking about the reals
    ##|x| < 1## means the open interval ##(-1,1)##
    ##|x| \leq s < 1## means the closed interval ##[-s,s]## for some ##0<s<1##.

    Hint: consider ##\frac{1+s}{2}##
  7. Dec 16, 2013 #6


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Why don't you tell us the context in which one is supposed to matter and the other doesn't. Typically it is relevant when you are talking about uniform convergence: the series
    [tex] \sum_{n\geq 0} x^n [/tex]
    converges in (-1,1) and converges uniformly in [-s,s] for any s < 1. So when doing proofs about integrating and differentiating the power series that require uniform convergence to hold you have to restrict yourself to [-s,s].
  8. Dec 18, 2013 #7
    I see no difference between the two either..
  9. Dec 18, 2013 #8
    I agree, you should show us the correct context.
  10. Dec 18, 2013 #9


    User Avatar
    Science Advisor

    "|x|< 1" refers to any number satisfying -1< x< 1. The set of such numbers includes numbers "arbitrarily" close to 1.

    "[itex]|x|\le s[/itex] for some s< 1" posits some specific bound on x less than 1. The set of such numbers does NOT include number "arbitrarily" close to 1.
  11. Dec 18, 2013 #10


    User Avatar
    Science Advisor
    Homework Helper

    The reason the two statements are different is: There is no real number that is the less than 1 but the "nearest" number to 1.

    A statement like yours would be correct if you were talking about integers. For example if N is an integer and ##|N| < 10##, then there is an integer S such that ##|N| <= S < 10##, and of course S = 9.

    But the same logic doesn't work for real numbers, or rationals, because you can always find another number in between ANY pair of real or rational numbers.

    Whatever number you choose for s, I can choose another number closer to 1 - for example (1+s)/2. So the number you call "some s" doesn't exist.

    This logical error is similar to the wrong idea that the number written as "0.999999......." in decimal notation is somehow different from the number "1".
  12. Dec 18, 2013 #11
    OK, so is the issue that we can't represent an open interval as a closed interval with different endpoints, because these endpoints would have to be, essentially, hyperreal?
  13. Dec 18, 2013 #12
    An open interval has no endpoints, not even hyperreal ones.
  14. Dec 18, 2013 #13
    Then replace my use of "endpoints" with "numbers that are next to the little parenthesis or bracket things in the notation" if you'd prefer.

    I don't know what you'd call x and y in the interval (x,y) besides endpoints. They aren't included in the interval, but I've never heard anyone hesitate to call them endpoints.
    Last edited: Dec 18, 2013
  15. Dec 18, 2013 #14
    I guess endpoint is a suitable name, but you seemed to imply that they belong to the interval, which they don't. You could always use suprema or infima.
  16. Dec 18, 2013 #15
    What I'm asking / saying is that from the replies above, if the OPs error is similar/equivalent to the idea that I cannot represent the interval:

    (x, y)

    As a closed interval

    [a, b]

    By defining new numbers a and b, because in order for the intervals to be equivalent, a and b would have to be defined in a way that made them not real numbers.
  17. Dec 18, 2013 #16

    Stephen Tashi

    User Avatar
    Science Advisor

    Technically those are not statements. To be a statement, you would have to give a complete context for those notations. For example, "There exists a number x such that |x| < 1". A statement must be a claim that is True or False.

    Interpreting |x| < 1 to mean "the set of numbers x such that |x| < 1" doesn't produce a statement. It produces a noun.

    Perhaps you mean to ask whether the two notations are notations for the same set of numbers. Interpreting |x| <1 as a specification of set defines a particular set. The second notation does not define a particular set unless there is more context that defines a particular s. Perhaps the context you are using is "There exists a number s such that s < 1 and the series converges for all numbers x such that |x| < s". That statement does not comment on whether the series converges for a number y such that s < y < 1.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook