Show that $C_{0} = \{(a_n)_{n \in \mathbb{N}}\subset \mathbb{R}: a_n \rightarrow 0\}$ is complete.

75 Views Asked by At

Show that $C_{0} = \{(a_n)_{n \in \mathbb{N}}\subset \mathbb{R}: a_n \rightarrow 0\}$ is complete.

I've already seen that this question has been asked, and already answered, however, I've managed to make a different proof and I wanted to check it was ok, since yesterday I had some trouble when trying to prove something diferent which involved a similar step.

Ok, so $C_0 \subset \mathcal{l^{\infty}}$ which I already proved its complete. Therefore, its enough to check that $C_{0}$ is closed. So, given $(x^{(n)}_{k})_{n \in \mathbb{N}}$ where for every $n\in \mathbb{N}$, $x^{(n)}_{k}=(x_{1}^{(n)},x_{2}^{(n)},...,x_{k}^{(n)},...)$. Lets say $(x^{(n)}_{k})\rightarrow (a_k)$ as $n\rightarrow \infty$. I want to see that $(a_k)\in C_0$ To conclude that $C_0$ is closed.

That is, given $\epsilon > 0$ I want to find $k_0 \in \mathbb{N}$ such that $|a_k|<\epsilon$ if $k \geq k_0$.

Now, since $(x^{(n)}_{k})\rightarrow (a_k)$, given $\frac{\epsilon}{2}>0$ there exists $n_1 \in \mathbb{N}$ such that $d_{\infty}((x^{(n)}_{k}),(a_k)) = sup_{k \in \mathbb{N}}|x^{(n)}_{k} - a_k|<\frac{\epsilon}{2}$ if $n \geq n_1$.

Now, fixing $n\geq n_1$, we have that given $\frac{\epsilon}{2}>0$ there exists $k_0 \in \mathbb{N}$ such that $|x^{(n)}_{k}|<\frac{\epsilon}{2}$ if $k\geq k_0$. Therefore we can say that

$|a_k|-|x^{(n)}_{k}|\leq |a_k-x^{(n)}_{k}| \leq sup_{k \in \mathbb{N}}|x^{(n)}_{k} - a_k|<\frac{\epsilon}{2}$.

Taking $k\geq k_0$

$|a_k|<\frac{\epsilon}{2}+|x^{(n)}_{k}|<\frac{\epsilon}{2}+\frac{\epsilon}{2}=\epsilon$

Concluding that $a_k \rightarrow 0$ and therefore, $a_k \in C_0$.

My only doubt about this proof is that my $k_0$ depends on the fixed $n\geq n_1$. But I don't see how that matters. (Im not sure if it does)!

1

There are 1 best solutions below

0
On BEST ANSWER

Your proof is perfectly fine, moreover, it's very similar to the usual proof of that fact. The dependence exists, but it doesn't affect the proof (look at $n$ as an additional parameter that you use to get a value smaller, and once you've done that, it's not needed anymore).

I'll make a suggestion: don't say "given $\frac{\epsilon}{2}>0$", I'd say instead "for $\frac{\epsilon}{2}>0$" or "since $\frac{\epsilon}{2}>0$".