Page 2 of 2 FirstFirst 12
Results 16 to 21 of 21

Math Help - continuety parameter question..

  1. #16
    Lord of certain Rings
    Isomorphism's Avatar
    Joined
    Dec 2007
    From
    IISc, Bangalore
    Posts
    1,465
    Thanks
    6
    If you tell me clearly, what part of my proof in post #8 and post #10, you have not understood, I will try to clarify it. Dont simply say Undefined * bounded etc

    Learn from your notes that testing for continuity is not just substituting the value at that point.
    Follow Math Help Forum on Facebook and Google+

  2. #17
    MHF Contributor
    Joined
    Nov 2008
    Posts
    1,401
    Quote Originally Posted by Isomorphism View Post
    That solution you got is of no use to part A. To demonstrate discontinuity at x_0, we have to find a sequence x_n \to x_0, but the sequence f(x_n) \not \to f(x_0).

    Your sequence x_n \to 0, but f(x_n) = \sin 2n \pi = 0. So it is of no use.

    As a side note, I am really sorry transgalactic, that I am failing in explaining this concept to you. I have already given you the answer for this in my previous posts. The sequence you can use is x_n = \frac2{(2n+1)\pi}

    I am telling the following based on our discussion.....If you take it in the right spirit, I suggest you take tutions. Sometimes a live tutor teaching a new concept is an invaluable experience.
    i cant see from where you came up with that sequence??
    x_n = \frac2{(2n+1)\pi}
    Follow Math Help Forum on Facebook and Google+

  3. #18
    Lord of certain Rings
    Isomorphism's Avatar
    Joined
    Dec 2007
    From
    IISc, Bangalore
    Posts
    1,465
    Thanks
    6
    Quote Originally Posted by transgalactic View Post
    i cant see from where you came up with that sequence??
    x_n = \frac2{(2n+1)\pi}
    Well I wanted x_n to converge to 0 and f(x_n) to not go to zero. I realised that trigonometric function alternates between +1 and -1, and for sin it happens when the angle is \frac{(2n+1)\pi}{2}. Thus I used it to my advantage. So by defining that sequence f(x_n) = (-1)^n, which does not have a limit.
    Follow Math Help Forum on Facebook and Google+

  4. #19
    MHF Contributor
    Joined
    Nov 2008
    Posts
    1,401
    ok i understood your last post
    regarding m=0 case
    i will try to prove what you asked
    Follow Math Help Forum on Facebook and Google+

  5. #20
    MHF Contributor
    Joined
    Nov 2008
    Posts
    1,401
    Follow Math Help Forum on Facebook and Google+

  6. #21
    MHF Contributor

    Joined
    Aug 2008
    From
    Paris, France
    Posts
    1,174
    Hi,

    I shall try to summarize the answers to your questions. You should also read Isomorphism's posts again from the beginning.

    For a given m\in\mathbb{N}, we consider the function f:x\mapsto \left\{\begin{array}{cc}x^m\sin\frac{1}{x} & \mbox{if }x\neq 0\\ 0 & \mbox{if }x=0\end{array}\right. defined on \mathbb{R}.

    First remark is that f, when restricted to the intervals (-\infty,0) or (0,+\infty), is infinitely differentiable (i.e. it can be differentiated any number of times). It is indeed obtained by product and composition of functions that are infinitely differentiable on their respective domain. And we can compute f'(x) for x\neq 0 using usual rules.

    It is not clear however what happens at 0 since \frac{1}{x} diverges when x\to0.

    a) The continuity at 0 means that f(x)\to f(0) when x\to 0.
    - Like you said, if m>0, then f is the product of a bounded function (the sine) by a function converging to 0 at 0, so that f converges to 0 at 0. Explicitly, we have |f(x)-f(0)|=|f(x)|\leq |x|^m and the right-hand side converges to 0 when x tends to 0. Thus f is continuous.
    - When m=0, we have f(x)=\sin\frac{1}{x} for x\neq 0. Plot this function on your calculator. When x goes down from 1 to 0, \frac{1}{x} goes up (continuously) from 1 to +\infty, so that \sin\frac{1}{x} oscillates infinitely many times (it is like the whole sine curve on [1,+\infty) were "compressed" (backwards) into the interval (0,1], with increasing compression rate on the neighbourhood of 0). Once you understand what the graph looks like, it is pretty obvious that f won't have a limit at 0. One reason is that f achieves the value 1 at points arbitrarily close to 0, whereas the continuity would mean that f is close to 0 on small neighboorhoods of 0. The rigorous proof goes through a sequence like the one introduced by Isomorphism: the points x such that f(x)=1 are such that \frac{1}{x}=\frac{\pi}{2}+2k\pi for some integer k. Thus, if x_k=\frac{1}{\frac{\pi}{2}+2k\pi}, we have f(x_k)=1 for all k, while obviously x_k\to_{k\to\infty} 0. If f was continuous at 0, we would have \lim_k f(x_k)=f(0)=0, whereas we have \lim_k f(x_k)=1: contradiction. Hence f is not continuous at 0.

    b) Differentiation at 0. Remember a function needs to be continuous in order to be differentiable. As a consequence, we only have to deal with the case m>0.
    f is differentiable at 0 iff the ratio \frac{f(x)-f(0)}{x-0} has a limit when x\to 0, x\neq 0. Note that this ratio equals \frac{f(x)}{x}=x^{m-1}\sin\frac{1}{x}, and m-1\in\mathbb{N}. Again, we know from the previous part that, at 0, this function converges to 0 if m-1>0 and it diverges if m-1=0. So that f is differentiable if and only if m>1, i.e. m\geq 2 since m is an integer. In addition, in this case, we have f'(0)=0 (the value of the limit).

    c) Continuity of the derivative at 0. The function needs to be differentiable, so we restrict to m\geq 2. First compute the derivative of x\mapsto x^m \sin\frac{1}{x} on \mathbb{R}\setminus\{0\}, like you usually do (Isomorphism did it). Then look at its limit at 0. The derivative is continuous iff this limit exists and is 0 (since we said that f'(0)=0). The derivative is found to be the sum of two terms that are pretty similar to f. One of them converges to 0 for any m\geq 2, while the other one converges to 0 iff m>2. Hence the sum converges to 0 iff m>2, i.e. m\geq 3.

    I hope this clarifies things in your mind...
    Follow Math Help Forum on Facebook and Google+

Page 2 of 2 FirstFirst 12

Similar Math Help Forum Discussions

  1. uuniform continuety question
    Posted in the Calculus Forum
    Replies: 0
    Last Post: September 3rd 2011, 02:11 PM
  2. integral/continuety proof question..
    Posted in the Calculus Forum
    Replies: 3
    Last Post: February 24th 2009, 11:16 AM
  3. continuety prove question..
    Posted in the Calculus Forum
    Replies: 3
    Last Post: December 27th 2008, 07:25 AM
  4. continuety of a complex function question..
    Posted in the Calculus Forum
    Replies: 1
    Last Post: December 25th 2008, 05:12 PM
  5. integral/continuety proof question..
    Posted in the Calculus Forum
    Replies: 0
    Last Post: December 21st 2008, 11:11 AM

Search Tags


/mathhelpforum @mathhelpforum