MHB Prove Riemann Integrability: Let f:[0,1]->R

  • Thread starter Thread starter mathster
  • Start date Start date
  • Tags Tags
    Riemann
Click For Summary
The discussion focuses on proving the Riemann integrability of a bounded function f:[0,1] that is continuous on (0,1). It is established that since f is continuous on the interval [t1, tn-1], it is uniformly continuous and thus integrable on that subinterval. A partition P is created to include endpoints, allowing the application of the uniform continuity property to show that the upper and lower sums can be made arbitrarily close. The argument is further strengthened by introducing a modified function g that matches f except at finitely many points, ensuring that g is also integrable. Ultimately, the conclusion is that f is Riemann integrable on [0,1] by demonstrating that the difference between the upper and lower sums can be made less than any given epsilon.
mathster
Messages
1
Reaction score
0
Let f:[0,1]-> R be bounded on [0,1] and continuous on (0,1). Prove that f is Riemann integrable on [0,1]. Hint: Show that for any epsilon > 0 there exists a partition P of [0,1] such that U(f,P)-L(f,P) < epsilon.

So Let P = {0 = t0 < t1 < ... < tn = 1}
Since its bounded on [0,1] (|f(x)| <= S) and continuous on (0,1) we know its continuous on [t1,tn-1]. So Let P1 = { t1, ..., tn-1} So its continuous on this interval and bounded we know that it is uniformly continuous and thus integrable on this interval so we know that U(f,P1) - L(f,P1) < epsilon for all epsilon. Now I guess I need to add in the sums of the two intervals on the end to make the partition equal to P. Am I on the right track? SO then I would have U(f,P) - L(f,P) = M(f,[t0,t1])*(t1-t0) + U(f,P1) + M(f,[tn-1,tn])*(tn-tn-1) - [ m(f,[t0,t1]*(t1-t0) + L(f,P1) + m(f,[tn-1,tn]*(tn - tn-1) ] <= S(t1-t0) + U(f,P1) + S(tn-tn-1) - [S(t1-t0) + L(f,P1) + S(tn-tn-1)] = U(f,P1) - L(f,P1) < epsilon.

I think this is true but also feel like I am missing something. Any help is much appreciated.
 
Last edited:
Physics news on Phys.org
mathster said:
So its continuous on this interval and bounded we know that it is uniformly continuous and thus integrable on this interval so we know that U(f,P1) - L(f,P1) < epsilon for all epsilon.

This is false. Carefully, write the sentence again using the right quantifiers.
---

Here is a hint. Suppose $f$ and $g$ are functions defined on $[0,1]$ where $f$ is integrable on $[0,1]$ and $g=f$ for every $x\in [0,1]$ except for finitely many values. Then $g$ is integrable also on this interval.

If $f:[0,1]\to \mathbb{R}$ is bounded consider $\lim_{x\to 0^+}f(x)$. This limit may not necessarily exist, so instead consider $\limsup_{x\to 0^+}f(x)$ which does exist and is equal to some finite number number $a$. In a similar manner let $b = \limsup_{x\to 1^-}f(x)$. Now define function $g = a$ at $x=0$, $g=b$ at $x=1$, and $g=f$ for $0<x<1$. By the above theorem to show that $f$ is integrable on $[0,1]$ it sufficies to show that $g$ is integrable.

Since $\limsup_{x\to 0+} g(x) = g(0) $ for any $\varepsilon > 0$ there is a $0<A < 1$ such that $|g(A) - g(0)| < \varepsilon$. Since $\limsup_{x\to 1^-} g(x) = g(1)$ there is a $A<B<1$ such that $|g(B) - g(1)| < \varepsilon$.

Now on the interval $[A,B]$ the function $g$ is uniformly continuous so we can partition it as $\{ t_1,t_2,...,t_{n-1} \}$ such that $t_1=A,t_{n-1}=B$ and $|g(t_j) - g(t_{j-1})| < \varepsilon$.

Define $P = \{ t_0,t_1,t_2,...,t_{n-1},t_n\}$ such that $t_0=0$ and $t_n = 1$. Then we see that $|g(t_j) - g(t_{j-1}) | < \varepsilon$. From here you should be able to carry out the rest of the argument.
 
Just take $$P=\{0,1/n,...,i/n,...,1\}$$, $$i=0,2,...,n$$.

Hence, $$f(x)$$ is uniformly continuous on $$[1/n, (n-1)/n]$$. Hence, $$M_i-m_i<\eta$$, for $$1\leq i\leq n-1$$ and large $$n$$.

Let $$M=\sup |f(x)|$$, we get $$M_i-m_i\leq 2M$$.

Therefore

$$U(P,f)-L(P,f)\\=\sum_{1\leq i\leq n-1}(M_i-m_i)/n+(M_0-m_0)/n+(M_n-m_n)/n\leq\eta(n-2)/n+4M/n <\epsilon$$

for large n.
 
We all know the definition of n-dimensional topological manifold uses open sets and homeomorphisms onto the image as open set in ##\mathbb R^n##. It should be possible to reformulate the definition of n-dimensional topological manifold using closed sets on the manifold's topology and on ##\mathbb R^n## ? I'm positive for this. Perhaps the definition of smooth manifold would be problematic, though.

Similar threads

  • · Replies 14 ·
Replies
14
Views
4K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 6 ·
Replies
6
Views
464
  • · Replies 2 ·
Replies
2
Views
2K
Replies
2
Views
2K