Suppose $f$ is Riemann integrable. Consider a sequence of partitions $P_n$ such that $mesh P_n$ tends to zero as $n \rightarrow \infty$. I want to prove that the sequence of upper sums converge i.e. $\lim_{n \rightarrow \infty}U(f,P_n)$ exists. How to do so ?

  • 1
    $\begingroup$ This could be the very definition of f being Riemann integrable. Please make precise which definition you are using. $\endgroup$ – Did Dec 16 '13 at 9:36

Hint: If $f:[a,b]\to \Bbb R$ a bounded function. Then $f$ is Riemann integrable in $[a,b]$ iff $\forall \epsilon >0$ there is a partition $P$ of $[a,b]$ such that $U(f,P)-L(f,P)<\epsilon$.

Use this in order to prove that $\lim_{n\to \infty} (U(f,P_n)-L(f,P_n))=0$.

Also $f$ is Riemann integrable iff $\underline \int _{a}^{b} f=sup_P(L(f,P))=inf_Q(L(f,Q))=\overline \int_{a}^{b} f$.

| cite | improve this answer | |
  • $\begingroup$ I understood till $\lim_{n\to \infty} (U(f,P_n)-L(f,P_n))=0$. But how does that prove what I want to prove ? $\endgroup$ – aaaaaa Dec 16 '13 at 11:15
  • $\begingroup$ @Prasenjit Because $f$ is Riemann integrable, $\lim_{n->\infty}U(f,P_n)$ will exist because of the last two lines that i wrote. $\endgroup$ – Haha Dec 16 '13 at 11:55
  • $\begingroup$ Can you elaborate ? $\endgroup$ – aaaaaa Dec 16 '13 at 13:16
  • $\begingroup$ I think this limit exists if the sequence $P_n$ is a increasing sequence (in terms of refinement), otherwise we can't say that the limit exists. My question is regarding that $\endgroup$ – aaaaaa Dec 16 '13 at 13:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.