0
$\begingroup$

Let $N\in\mathbb{N}$ be a natural number and let $\alpha_0,\alpha_1,...,\alpha_N$ be real numbers such that $\alpha_i\neq\alpha_j \forall i\neq j$

We define in $X$ (the vector space of polynomials of degree $\leq N$) the following: $$ \lVert\mathcal{P}\rVert = \sum_{k=0}^N\lvert \mathcal{P}(\alpha_k)\rvert, (\mathcal{P}\in X) $$ To prove:

  1. $\lVert .\rVert$ is a norm.
  2. The topology generated by this norm does not depend on the $\alpha_k$.
  3. A sequence $\{\mathcal{P}_n\}$ in X converges with the norm $\lVert.\rVert_\infty$ ($\lVert f\rVert_\infty=sup\{\lvert f(t)\rvert: t\in[a,b]\}$) in $[a,b]$ if, and only if, there exists $\beta_0,\beta_1,...,\beta_N\in[a,b]$ with $\beta_i\neq\beta_j \forall i\neq j$ such that $\{\mathcal{P}_n(\beta_k)\}_{n\in\mathbb{N}}$ converges $\forall k=0,1,...N$.

The first and second points are trivial: proving that $\lVert.\rVert$ is a norm consists only in applying the definition and, since all the norms are equivalent in a finite dimensional vector space, the topology generated does not depend on the $\alpha_k$.

My problem reaches with the third point: do I have to use that the convergence of $\{\mathcal{P}_n\}$ is equivalent to the convergence of its coordinates?

Could you please give me any hint to solve this problem?

Thank you.

$\endgroup$
  • 1
    $\begingroup$ The nomr you mention in (3) is not the norm you defined...is it? I mean, because you wrote $\;||.||_\infty\;$ ... $\endgroup$ – DonAntonio Feb 19 '14 at 16:55
  • $\begingroup$ Yes, @DonAntonio, it is different. I edited the post explaining it, sorry. $\endgroup$ – Alejandro Feb 19 '14 at 16:59
0
$\begingroup$

I thought that I could use that, since all the norms are equivalent in any finite dimensional vector space, the convergence with the norm $\lVert.\rVert_\infty$ is equivalent to the convergence with the norm $\lVert.\rVert$, defined in 1.

This way, I could argue that the convergence with $\lVert.\rVert$ is equivalent to the convergence of every addend in the summation (if only one addend does not converges, the summation does not converges).

It's quite intuitive and I don't know if it is rigorous, what do you think?

$\endgroup$
  • $\begingroup$ Any opinion? I am not completely sure of this. $\endgroup$ – Alejandro Feb 19 '14 at 17:47

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.