I'm reading the book Real Analysis of Folland. When I reached chapter 2 about the different modes of convergence, there's an example Folland gave that confused me:

The 2 function sequences:
i) $f_n = \chi_{(n, n+1)}$
ii) $f_n = n\chi_{[0, 1/n]}$
uniformly, pointwise converge to $0$, but not converge in $L^1$

I can easily point out that those 2 functions pointwise converge to $0$ as well as not converge in $L^1$. But I don't think those 2 functions uniformly converge. Like the first case, for all $N$ and choose $\epsilon < 1$, we can always choose $x \in (N , N+1)$, then $|f_n(x)| = 1 > \epsilon$. Did I miss some points? (The example is in chapter 2, page 61) Please help me to judge this. Thanks so much.


You misunderstood what he says. Folland writes:

(i) $f_n=n^{-1}\chi_{(0,n)}$

(ii) $f_n=\chi_{(n,n+1)}$

(iii) $f_n=n\chi_{[0,1/n]}$

In (i), (ii), and (iii), $f_n\to 0$ converges uniformly, pointwise, and a.e. respectively.

So $f_n=\chi_{(n,n+1)}$ does not converge uniformly, but pointwise which is evident. Only the first one converges uniformly.

  • $\begingroup$ Ah yeah, so it means the first case converges uniformly, the second pointwise and the last a.e, right. Thanks so much for clarifying. $\endgroup$ – le duc quang Jul 23 '15 at 2:28

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.