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!

Homework Help: Integral of open set

  1. Dec 3, 2012 #1
    1. The problem statement, all variables and given/known data


    3. The attempt at a solution
    for part A i assume that f is in norm space C[0,1],||.|| , then choose a sequence fn in C[o,1] s.t fn->f then for 0<fn<1 so 0<f<1 i.e. A is closed i am not sure my answer here

    for part B i assume the anti-derivatice of f(t) to be K(t)+c therefore, by F(f)=2K(1/2)+1/2-K(1)-K(0) then how should i prove it is cts ???
  2. jcsd
  3. Dec 3, 2012 #2
    I don't see how that proves anything. You need to prove that if [itex]f_n[/itex] is a convergent sequence in A such that [itex]f_n\rightarrow f[/itex], that then [itex]f[/itex] is an element of A.

    OK, but that doesn't really help. What definition of continuous would you like to use here? Can you state it?

    Also, it would be a great help for us if you would type your posts in LaTeX: https://www.physicsforums.com/showpost.php?p=3977517&postcount=3
  4. Dec 3, 2012 #3
    if [itex]f_n[/itex] is a convergent sequence in A such that [itex]f_n\rightarrow f[/itex], then for [itex]f_n[/itex] is in A , A is a subset of C[0,1] so fn is bounded by 0 and 1 and continuous on [0,1] so fn is uniformly continuous to f ,then f is bouded by 0,1 which is in A????

    and secound part if i use the definition of continuous , then exists a δ,s.t. |f(x)-f(y)|<δ implies that |F(f(x))-F(g(x))|<esillope . and just fixed the esillope . and i can solve it out now many thanks
    Last edited: Dec 3, 2012
  5. Dec 3, 2012 #4
    Sure. But can you actually prove that f is bounded by 0 and 1? Once you proven that, I agree that the above shows that A is closed.

    Protip: we usually talk about epsilon instead of esillope

    OK, so you you need to make [itex]|F(f)-F(g)|[/itex] small somehow by making [itex]\|f-g\|_\infty[/itex] small. So, let us start by writing

    [tex]|F(f)-F(g)|=\left|\int_0^{1/2}f(t)dt + \int_{1/2}^1 (1-f(t))dt - \int_0^{1/2} g(t)dt - \int_{1/2}^1 (1-g(t))dt\right|[/tex]

    Try to find a good estimation for that.
  6. Dec 3, 2012 #5
    i not sure how to prove that 0<f<1 ,which way should i begin with??
  7. Dec 3, 2012 #6
    You should not prove 0<f<1, you should prove [itex]0\leq f\leq 1[/itex].

    Start by fixing an [itex]x\in [0,1][/itex]. Then try to deduce from [itex]f_n(x)\rightarrow f(x)[/itex] that [itex]0\leq f(x)\leq 1[/itex].
  8. Dec 3, 2012 #7
    for me ,it is obviouse. [itex]0\leq f_n\leq 1[/itex]. then max fn =1, min fn =0 and limfn =f so by sanwich rule [itex]0\leq f(x)\leq 1[/itex] i just dont know how should i prove that f is bounded by [0,1] here
  9. Dec 3, 2012 #8
    Why should [itex]\max f_n=1[/itex] and [itex]\min f_n=0[/itex]?? That's really weird.
    For example, take the map

    [tex]f_n:[0,1]\rightarrow \mathbb{R}:x\rightarrow x\rightarrow \frac{1}{n}[/tex]

    then the [itex]\max f_n=\min f_n=\frac{1}{n}[/itex].

    You're really making it harder than it is. I'm just asking you that if [itex](y_n)_n[/itex] is a convergent sequence with [itex]y_n\rightarrow y[/itex], and if [itex]0\leq y_n\leq 1[/itex], then [itex]0\leq y\leq 1[/itex]. It doesn't even have anything to do with functions.
  10. Dec 3, 2012 #9
    omg! thats what i thought at the beginning!!! many thanks !!!
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook