Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $\Vert \cdot \Vert$ a norm on $X=C([0,1])$ s.t.

1. $X$ is complete w.r.t. $\Vert \cdot \Vert$;

2. convergence in $\Vert \cdot \Vert$ implies pointwise convegence, i.e. $$ \Vert x_n - x \Vert \to 0 \Rightarrow \forall t \in [0,1], \quad x_n(t) \to x(t). $$

Is $\Vert \cdot \Vert$ equivalent to the usual $\sup$-norm, $\Vert \cdot \Vert_{\infty}$?

By open mapping and 1 (completeness), we just need to prove only one inequality of the two we need to prove equivalence (indeed, the other would follow by open mapping).

Anyway, I think the answer is affirmative; am I right? I do not know how to prove it. Any ideas, please? Thanks.

share|improve this question
2  
Perhaps this is easy from the closed graph theorem... –  GEdgar Dec 31 '12 at 18:19
    
A variation on the same theme. In particular, by perturbing the $\lVert\, \cdot\, \rVert_\infty$-norm with a discontinuous functional it is shown that one can have point 2. for an inequivalent norm. –  Martin Dec 31 '12 at 19:41

1 Answer 1

up vote 5 down vote accepted

The hypothesis implies that for all $t\in I=[0,1]$ the linear functional $\mathrm{ev}_t:X\to\Bbb R, ~f\mapsto f(t)$ is continuous. Also, for any $f\in X$, $$\sup_{t\in I}~|\mathrm{ev}_t(f)|=||f||_{\infty}<+\infty$$ Since $X$ is a Banach space, the uniform boundedness principle applies, and $$ C=\sup_{t\in I}~||\mathrm{ev}_t||_{X'}<+\infty$$ that is, for any continuous function $f$ and any $t\in I$, $|f(t)|\leq C||f||_X$ thus $$||\cdot||_{\infty}\leq C||\cdot||_X$$ As you noted, the open mapping theorem then tells us that both norms are equivalent.

share|improve this answer
    
Thanks, nice application of uniform boundedness. Very useful, thanks again for your help. –  Romeo Dec 31 '12 at 19:43
    
My pleasure Romeo : ) happy new year btw –  Olivier Bégassat Dec 31 '12 at 19:53
    
Happy new year to you, too :-) –  Romeo Dec 31 '12 at 19:54

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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