The Limit of a Sequence is Unique Proof

Поділитися
Вставка
  • Опубліковано 24 гру 2024

КОМЕНТАРІ • 67

  • @londynocchipinti6607
    @londynocchipinti6607 2 роки тому +13

    truly do not understand why professors make this so complicated. the way you explain it is perfect! thank you!

    • @TheMathSorcerer
      @TheMathSorcerer  2 роки тому

      You're very welcome!

    • @enkixloki
      @enkixloki Рік тому +1

      pls someone explain to me what does episilon,N,n....represent...im still confused

  • @debendragurung3033
    @debendragurung3033 5 років тому +19

    Its been a month I am scratching my head around this proof I have on a lecture note. And it's been 4 years since this video been uploaded. I am glad I found it finally and I can actually make a good sense of it now. You are a saviour mate. Thanks and keep them coming.

    • @TheMathSorcerer
      @TheMathSorcerer  5 років тому +3

      So glad it helped this stuff is hard😄

    • @enkixloki
      @enkixloki Рік тому

      pls someone explain to me what does episilon,N,n....represent

    • @debendragurung3033
      @debendragurung3033 Рік тому +1

      @@enkixloki You have to understand from the very beginning and also understand why, why is the bigger question.
      First thing is we are using Natural number tools to understand real numbers. That's the reason why we see use of Natural numbers everywhere throughout this lecture.
      We have to assume we know very little about real numbers. That's even though we have been pretaught many things about real numbers before college and we have been using them without evidence of concrete proof.
      The best place to start as any lecturers teach is the existence of numbers that are not rational. First of course in the liat is sqrt(2).
      Earlier mathematician did believe that any numbers between two integers must be some fractional number . You can check an easy proof that sqrt(2) cannot be a fraction. Yet simple logic tells u it must exist and it is lesser than 2 but greater than 1. So what other numbers are there in the interval [1,2] that are not rational. Hence the irrational.
      Now what does sequence have to do with any of this. To define sequence we use NATURAL NUMBERS. It's the tool we are all born with. We can discretely count them 1,2,3.... And we can without any proof say 3 is greater than 2 and 2 is greater than 1. We are simply using this tool to reconstruct understanding of real numbers.
      We can of course never write out all real numbers using this fixed set of symbols in natural numbers. But best we can do is get very close to it. For instance π =3.14...... with endless digits of natural numbers. What we doing here is simply getting it's approximate value. 3

    • @enkixloki
      @enkixloki Рік тому

      @@debendragurung3033 i appreciate your explanation but im still stuck with it i will try to comprehend the topic more....i would like it if u helped me more it seems like u have understand it well and i wanna reach your level just show me z way

  • @smoosq9501
    @smoosq9501 3 роки тому +2

    my god!!!!, i was really confuse when the book only said the |a-b|

  • @jmitchell8082
    @jmitchell8082 3 роки тому +2

    Thank you for this, so much clearer than my lecture notes

  • @TheMathSorcerer
    @TheMathSorcerer  9 років тому +6

  • @shirleymoon9934
    @shirleymoon9934 Рік тому +2

    Very helpful 👍🏻

  • @successorof68
    @successorof68 2 роки тому +4

    Its been 7 years since you uploaded, no way you are going to reply but worth a shot.
    how did you get eps/2+eps/2 from |a-an|+|an-b|? really hope you see this

    • @TheMathSorcerer
      @TheMathSorcerer  2 роки тому +5

      Each of those is less than eps/2 so i replaced each with eps/2

  • @aaryadhurwey6282
    @aaryadhurwey6282 Рік тому +6

    Why we had written there (an-a)less than epsilon by 2??

    • @covariance5446
      @covariance5446 9 місяців тому +1

      If Xn converges to some value (call it a) then by definition you can say that when you have a certain "roadmarker" (call it N or N-nought or N-sub-1, the name doesn't matter) in the element of the naturals, it will be the case that the absolute value of the difference between Xn and a should be able to be smaller than any number you choose. Traditionally, the choice is to call that value Epislon, but it's just name, and you can call it whatever you want. If you're just trying to say that the abs.value of the difference between xn and a can be made arbitrarily small, you can also say that as n progresses past your roadmarker, then the diff between xn and a will also be smaller than epislon/two.
      The nice thing about doing this is so that when he (Math Sorc) sets up the inequality, you're adding two terms for which each term is an absolute difference between xn and the ostensible limit.
      If each one was said to be less than epislon over two, then the sum of both can also be said to be less than 2*epislon/2. Just like how if 1 < 5 and 2 < 5, then 1+2 is less than 5 + 5.
      So, now you have the sum of |a-xn| + |xn-b| being less than epsilon.
      This allows us to explicate the contradiction (of |a-b| being less than epislon, which means |a-b| being less than |a-b| which is an obvious contradiction).

  • @louismensah4700
    @louismensah4700 2 роки тому +8

    Why e/2 ?

  • @Hosseinshafii
    @Hosseinshafii 6 років тому +13

    why epsilon/2 ?

    • @joaohax52
      @joaohax52 4 роки тому

      well, since you can set every epsilon within, epsilon/2 is one of the possibilities. that's just a step to show that there is an epsilon that doesnt follow these conditions (you can show that there are more than just one, but it is a different way..

  • @azhinrafiq4676
    @azhinrafiq4676 4 роки тому +1

    Thanks 👍🏻

  • @ROCKaholic
    @ROCKaholic 6 років тому +1

    What about delta (δ)? What if there is some positive delta that lives inside the Reals that exists in the definition of a limit? My book defines a limit as such:
    (Given a point of interest, a possible limit value, a defined function f: S -> R* where R* means the set of all real numbers and S ⊆ R* contains a deleted neighborhood of a)
    The limit of f(x) as x approaches a = L is defined as:
    IF [∀ (positive) ε ∈ R*, , ∃ (positive) δ ∈ R* such that ∀ x ∈ S; 0 < |x - a| < δ], then | f(x) - L | < ε.
    You didn't seem to include this type of definition, which is fair because every class and every book teaches it differently and your method is both true and helpful, but I learned it a little differently, so with what I gave, how could this proof be changed to pander to this formal definition of a limit?

    • @joaohax52
      @joaohax52 4 роки тому +1

      we need to realise that an and bn already converge to a and b, respectively. it doesnt need to be mentioned, since we just need to find an epsilon such that doesnt follow these condirions and then shows that is false (for every case)

    • @joaohax52
      @joaohax52 4 роки тому +1

      and if a = b, |a-b| < epsilon => 0 < epsilon, it returns into a true, that always epsilon is positive

  • @shreyashraj6906
    @shreyashraj6906 4 роки тому +1

    Thanku🙂

  • @khadijaeldesouky4586
    @khadijaeldesouky4586 4 роки тому

    i read all the comments about why is it epsilon divided by 2 and why is epsilon equal to abs(a-b) but i still can't understand.

    • @korinaldamtajima
      @korinaldamtajima 2 роки тому

      I hope you got it but in case you didn't and for anyone else stuck, read the section 3 "common questions" part in my Real Analysis progress diary blog: rakugakistudy.com/tutorials/62708b581b716f001650bf0b
      Hopefully it helps :)

  • @MrBABYCOLZ
    @MrBABYCOLZ 7 років тому +12

    Why is it epsilon/2?

    • @doodledeedah
      @doodledeedah 7 років тому +5

      Probably a little late for your exam, but you may think of ε as (a-b)/2 it's the same thing and clearer.
      i.e Suppose a != b. W.l.o.g. we may assume that a < b. We choose ε = (|a-b|)/2 > 0 (as a is bigger than b). Our epsilon is half the distance between a and b shown below.
      a a+ε = b - ε b
      ^ ^ ^
      Here a is one limit, b is the other.
      Since (an)n∈N converges to a, there is some N1 ∈ N such that
      a − ε < an < a + ε for all n ≥ N1 - this is the interval our limit for an lies in.
      Since (an)n∈N converges to a, there is some N2 ∈ N such that
      b − ε < an < b + ε for all n ≥ N2 - this is the interval our other supposed limit for an ALSO lies in.
      hence using n ≥ N1 and n ≥ N2
      take the lower bound of the limit b's interval and the upper bound of limit a's interval we get
      b - ε < an < a - ε < ---- this is a contradiction since a+ε = b - ε must hold for our initial assumption a != b to be true.
      Therefore our initial assumption is false and limit a == b (limit is unique QED)

    • @kendedetar
      @kendedetar 5 років тому

      @@doodledeedah You stated that a < b, then said "(as a is bigger than b)". So you should either write |b-a|/2 = ε, or just assume that a > b. But correct me if I'm wrong :)

    • @joaohax52
      @joaohax52 4 роки тому

      @@kendedetar |a-b| = |b-a| since it's always positive

  • @nmtran
    @nmtran 8 років тому

    Thank you so mucchh

  • @msk.nfazal9757
    @msk.nfazal9757 5 років тому +1

    Thanks :-)

  • @lemyul
    @lemyul 5 років тому +2

    it doesnt, but it will, later, hopefully

  • @kalaivani.v9833
    @kalaivani.v9833 6 років тому +1

    Tq

  • @mohmd252
    @mohmd252 9 років тому +2

    oh sorry , your proof is correct, please forgive me

  • @lijinnan9647
    @lijinnan9647 5 років тому

    great!

  • @SequinBrain
    @SequinBrain 3 роки тому +3

    I don't get these, but I'll keep trying anyway.

    • @TheMathSorcerer
      @TheMathSorcerer  3 роки тому +1

      Yeah keep trying man,I should and will make more of these!!

    • @DouaaZerarga-t1u
      @DouaaZerarga-t1u Місяць тому +1

      I'm reading ur comment now after three years u wrote it nd it really motivated me to keep going 😭😭😭😭😭😭

  • @GagandeepKaur-mw4ih
    @GagandeepKaur-mw4ih 7 років тому +4

    pls tell me why is it epsilon by 2

    • @doodledeedah
      @doodledeedah 7 років тому +2

      Probably a little late for your exam, but you may think of ε as (a-b)/2 it's the same thing and clearer.
      i.e Suppose a != b. W.l.o.g. we may assume that a < b. We choose ε = (|a-b|)/2 > 0 (as a is bigger than b). Our epsilon is half the distance between a and b shown below.
      a a+ε = b - ε b
      ^ ^ ^
      Here a is one limit, b is the other.
      Since (an)n∈N converges to a, there is some N1 ∈ N such that
      a − ε < an < a + ε for all n ≥ N1 - this is the interval our limit for an lies in.
      Since (an)n∈N converges to a, there is some N2 ∈ N such that
      b − ε < an < b + ε for all n ≥ N2 - this is the interval our other supposed limit for an ALSO lies in.
      hence using n ≥ N1 and n ≥ N2
      take the lower bound of the limit b's interval and the upper bound of limit a's interval we get
      b - ε < an < a - ε < ---- this is a contradiction since a+ε = b - ε must hold for our initial assumption a != b to be true.
      Therefore our initial assumption is false and limit a == b (limit is unique QED)

    • @timechanges1670
      @timechanges1670 5 років тому

      Bcz v hv to prove tat epsilon so dat v r taking epsilon/2

  • @enkixloki
    @enkixloki Рік тому

    pls someone explain to me what does episilon,N,n....represent

    • @TechToppers
      @TechToppers Рік тому

      Look up the definition of convergence of a sequence.

  • @jisungyu2
    @jisungyu2 6 років тому +1

    How did you come up with epsilon = abs (a-b)? what if epsilon is not equal to abs(a-b)?

    • @DOtman2222
      @DOtman2222 6 років тому +8

      In the definition of the convergence of the sequence:
      "For ANY epsilon >0,.."
      this means whatever positive number you pick for epsilon the other conditions should work(so that this sequence is convergent).
      Now this allows us to pick any number we want for epsilon(and the conditions will remain true) therefore we pick
      |a-b| which will help in attaining contradiction.hope that helps

    • @Codepmz.
      @Codepmz. 5 років тому

      @@DOtman2222 well replied

  • @suchetasinha2205
    @suchetasinha2205 5 років тому

    Sir why we r taking the max value of n1 and n2 ....????

    • @TheMathSorcerer
      @TheMathSorcerer  5 років тому +3

      well we need n > N_1 and n > N_2, so if we let M be the maximum, whenever n > M, it's also bigger than N_1 and N_2, so we have what we want(this way both inequalities hold)

  • @Sureshiru
    @Sureshiru 4 роки тому

    Why u take Max{N1,N2} value

  • @soumyaranjanbiswal1827
    @soumyaranjanbiswal1827 5 років тому +6

    Why r u taking€/2

    • @elisha2358
      @elisha2358 4 роки тому +1

      Because we assume (an) converges to 2 different values, namely, a and b.

  • @ethanperry9544
    @ethanperry9544 Рік тому

    great proof and all but you should use Natural number instead of Z+😠

  • @alexandertownsend3291
    @alexandertownsend3291 4 роки тому +1

    Unique proof of what? Lol

    • @tracyjannalzaro5099
      @tracyjannalzaro5099 4 роки тому +1

      He meant, "The Limit of a Sequence is Unique, Proof." hhe. :)

  • @twebazegilbert6871
    @twebazegilbert6871 7 місяців тому

    You are actually complicating math😢