How to prove a cubic equation has exactly one root on an interval (IVT, Contradiction, Rolle's Thm)

แชร์
ฝัง
  • เผยแพร่เมื่อ 18 พ.ย. 2024

ความคิดเห็น • 22

  • @Ninja20704
    @Ninja20704 17 ชั่วโมงที่ผ่านมา +14

    It would have been helpful to at least state the Rolle’s theorem.
    For those who don’t know, the theorem states that if a function f is continuous on [p,q], differentiable on (p,q) and f(p)=f(q) (need not be 0), then there exists r in (p,q) such that f’(r)=0
    Here we are using [a,b] as our interval, which is a sub-interval of [-2,2] so the continuity and differentiability requirements still hold for [a,b].
    Good video nontheless

    • @LuisFernando-wm5rg
      @LuisFernando-wm5rg 16 ชั่วโมงที่ผ่านมา

      Thx!

    • @dan-florinchereches4892
      @dan-florinchereches4892 11 ชั่วโมงที่ผ่านมา

      I don't think Rollers theorem itself will be very useful but the consequences:
      A) between 2 consecutive roots of f(X)=0 there will be at least one c such that f'(c)=0 between them
      B) between two consecutive roots of f'(X)=0 there will be at most 1 root of f(X)=0
      Or we could use the consequences of theorem of Laplace stating that if f'(X)

  • @ghstmn7320
    @ghstmn7320 11 ชั่วโมงที่ผ่านมา +3

    you can also do it using monotony after IVT. f is differentiable as a polynomial function in the interval [-2,2] with a derivaive f'(x)=3x^2 -15. After simple computation its proven that f'(x)

    • @dlevi67
      @dlevi67 10 ชั่วโมงที่ผ่านมา +2

      That (your first reasoning line - it's continuous and strictly decreasing, thus it has no other zero by IVT) is how I approached it too.

  • @ianfowler9340
    @ianfowler9340 12 ชั่วโมงที่ผ่านมา +5

    After showing there is at least one root in the interval [2,2], I think it is sufficient to show that there are no local max/min in [2,2] as that prevents any more roots. It's a similar argument but easier to present. Am I missing something?

    • @dlevi67
      @dlevi67 10 ชั่วโมงที่ผ่านมา +1

      Don't know - possibly Steve wanted to use Rolle's theorem to show a use case.

    • @bprpcalculusbasics
      @bprpcalculusbasics  6 ชั่วโมงที่ผ่านมา +1

      I think showing there are no local min/max on [-2, 2] is equivalent to what I did. The reason that I used the Rolle's theorem is because that's just the book's way.... (Single Variable Calculus, by Stewart, sect 4.2) I know it's not the best reason, haha.

  • @henrymarkson3758
    @henrymarkson3758 15 ชั่วโมงที่ผ่านมา +5

    Isn't the IVT and f'(x)

    • @ianfowler9340
      @ianfowler9340 10 ชั่วโมงที่ผ่านมา +1

      I think so.

    • @Grecks75
      @Grecks75 9 ชั่วโมงที่ผ่านมา +2

      Yes, it is sufficient. And bprp effectively used that in his proof by contradiction by showing that f'(x) < 0 on the interval (-2, 2). Since the function f is strictly decreasing on this interval, it cannot have more than one zero in it.

  • @Albaraa
    @Albaraa 11 ชั่วโมงที่ผ่านมา

    Hey what’s square root of y(x) = x^2
    Calculate that then use x=3

  • @amoghopprasad8286
    @amoghopprasad8286 15 ชั่วโมงที่ผ่านมา +1

    what if lets say the intervals are different numbers and not have the same absoulte value

    • @bjornfeuerbacher5514
      @bjornfeuerbacher5514 11 ชั่วโมงที่ผ่านมา +1

      Doesn't change the way one solves this.

  • @christoskettenis880
    @christoskettenis880 15 ชั่วโมงที่ผ่านมา +1

    This method is widely used in the Year 12 elective mathematics course in Cyprus. Here's a link to a video of mine using the exact same technique: th-cam.com/video/eFtM8_wLpwc/w-d-xo.html
    Use generated-subs as the video is in Greek

    • @dan-florinchereches4892
      @dan-florinchereches4892 10 ชั่วโมงที่ผ่านมา

      Hi there. Kind of unrelated to the post but I am visiting Paphos is there something nice I shouldn't miss while I am here.
      Kind regards

  • @mr_angry_kiddo2560
    @mr_angry_kiddo2560 16 ชั่วโมงที่ผ่านมา

    What is limit of ((x-2)(x-3)/(x-4)(x-5)) as x tends to 5.
    Answer:zero(0) .
    I need mathematical proof 😢

    • @Neun_owo
      @Neun_owo 15 ชั่วโมงที่ผ่านมา +1

      The answer shouldn't be 0, as the limit doesn't exist!
      For a formal proof, we'll have to use Epsilon-Delta proof. but for the sake of simplicity, lets just say that the limit doesn't exist because when x tends to 5 either from left or right side. (x-5) in the denominator becomes incredibly small, so the function shoots up to infinity as it is trying to divide by a very small number. Thus the limit doesn't exist as x tends to 5.

    • @michaelfaccone5811
      @michaelfaccone5811 10 ชั่วโมงที่ผ่านมา +1

      ​@@Neun_owoone small correction, though I agree with your result. The limit as x approaches 5 from the left is actually negative infinity. The limit approaching from the right, as you said, is positive infinity. Since the two limits are different, then as you said the overall limit does not exist.

    • @Neun_owo
      @Neun_owo 10 ชั่วโมงที่ผ่านมา

      @@michaelfaccone5811 Ty for the correction!

    • @dlevi67
      @dlevi67 10 ชั่วโมงที่ผ่านมา +1

      Is the denominator just (x-4), i.e. the fraction is (x-3)/(x-4), or is the denominator (x-4)(x-5)?
      If the latter, then the two responses above are correct; the denominator will go to zero as x -> 5. However, I think your exercise has only (x-4) in the denominator, which means you have three factors in the expression:
      (x-2),
      (x-3)/(x-4)
      (x-5)
      for the limit to be zero, one of those three has to -> 0 for x -> 5 and the other factors need to exist in the same neighbourhood. You can easily demonstrate (with ε-Δ if rigorous proof is required; by inspection and noting that all monomials are continuous and differentiable in R if a less formal proof is adequate) that (x-5) -> 0 if x -> 5, and all the other factors tend to a finite value, thus their product (i.e. the whole expression) tends to 0.

  • @ziplock007
    @ziplock007 9 ชั่วโมงที่ผ่านมา

    Dumb question: why not just solve for the roots using the cubic equations and showing there's only one between the interval????