3

I need to prove that if $f$ is Riemann integrable $[0,1]$ then $$\lim\limits_{n\rightarrow\infty} \frac{1}{n}\sum f\left(\frac{k}{n}\right)= \int\limits_{0}^{1} f(x)dx$$

My idea is to recognize the right side defining the limit as a Riemann sum and use the uniform continuity.

I am not sure what to do next

Thanks

rtybase
  • 16,907
Pol
  • 369
  • 3
  • 11
  • Yes, your idea should work. Start by splitting [0,1] into $n$ segments of length $\frac{1}{n}$ each and use this as a partition. – rtybase Jan 28 '17 at 23:25
  • 2
    If you're assuming that $f$ is Riemann integrable, then you shouldn't need uniform continuity—just recognizing that the expression inside the limit on the left-hand side is a Riemann sum should be enough. – Greg Martin Jan 29 '17 at 00:24
  • 2
    What's your definition of Riemann integrability? – zhw. Jan 29 '17 at 00:26
  • @rtybase, could you please explain it more? I started to prove it with 1/n<ẟ. Then I try to express this : 1/n∑f(k/n)-∫fdx – Pol Jan 30 '17 at 23:14

1 Answers1

4

Take the following partition $$x_0=0<\frac{1}{n}<\frac{2}{n}<\frac{3}{n}<...<\frac{n-1}{n}<x_n=1$$ or $x_k=\frac{k}{n}$.

Now, the left Riemann sum is $$\sum_{k=1}^{n}f(x_{k-1})(x_k-x_{k-1})=\sum_{k=1}^{n}f\left(\frac{k-1}{n}\right)\left(\frac{k}{n}-\frac{k-1}{n}\right)=\frac{1}{n}\sum_{k=1}^{n}f\left(\frac{k-1}{n}\right)$$ And the right Riemann sum is $$\sum_{k=1}^{n}f(x_{k})(x_k-x_{k-1})=\sum_{k=1}^{n}f\left(\frac{k}{n}\right)\left(\frac{k}{n}-\frac{k-1}{n}\right)=\frac{1}{n}\sum_{k=1}^{n}f\left(\frac{k}{n}\right)$$ Combining this with the fact that $f$ is Riemann integrable (this is given) the limits of both left and right sums exist and are equal to $\int_{0}^{1}f(x)dx$. To understand this properly, it is important to understand the definition of Riemann integrability, which operates with "For $\forall \varepsilon >0$, there $\exists\delta$ such that for any partition ...", which also includes the one constructed above.

rtybase
  • 16,907