I am trying to prove this statement about continuity on the reals:]

Let $f:\mathbb R\to\mathbb R$ be continuous using the Euclidean distance function. Then the function $g:\mathbb R^n \to\mathbb R^n$ defined by $$g(x_1,x_2,...,x_n)=(f(x_1),f(x_2),...,f(x_n))$$ is continuous in Euclidean $n$-space.

Here is my proof:

Since $f$ is continuous, for any given value of $\epsilon\gt 0$, there exists $\delta\gt 0$ such that $$|a-b|\lt \delta\implies|f(a)-f(b)|\lt \epsilon$$ I will use $\delta[\epsilon]$ to represent such a value of $\delta$.

Now suppose that $\epsilon'\gt 0$ is given. Let $$\delta'=\delta\bigg[\frac{\epsilon'}{\sqrt n}\bigg]$$ Then suppose we have that $$\sqrt{\sum_{k=1}^n (a_k-b_k)^2}\lt \delta'$$ $$\sum_{k=1}^n (a_k-b_k)^2\lt \delta'^2$$ and, for each $k$, $$(a_k-b_k)^2\lt \delta'^2$$ $$|a_k-b_k|\lt \delta'$$ $$|a_k-b_k|\lt \delta\bigg[\frac{\epsilon'}{\sqrt n}\bigg]$$ and, by the way we defined $\delta[\epsilon']$, this implies $$|f(a_k)-f(b_k)|\lt \frac{\epsilon'}{\sqrt n}$$ $$n(f(a_k)-f(b_k))^2\lt \epsilon'^2$$ $$(f(a_k)-f(b_k))^2+(f(a_k)-f(b_k))^2+...+(f(a_k)-f(b_k))^2\lt \epsilon'^2$$ where the LHS of the above statement contains $n$ copies of $(f(a_k)-f(b_k))^2$. Since this is true for all $k$, even the $k$ for which $(f(a_k)-f(b_k))^2$ is maximized, we have that $$(f(a_1)-f(b_1))^2+(f(a_2)-f(b_2))^2+...+(f(a_n)-f(b_n))^2\lt \epsilon'^2$$ $$\sum_{k=1}^n (f(a_k)-f(b_k))^2\lt \epsilon'^2$$ $$\sqrt{\sum_{k=1}^n (f(a_k)-f(b_k))^2}\lt \epsilon'$$ Which completes the proof.

Is this proof correct?

  • 2
    $\begingroup$ Correct. However, the remark about $(f(a_k) - f(b_k))^2$ maximization is unnecessary, just add up the inequalities $(f(a_k)-f(b_k))^2 < \epsilon'^2/n$ for $k=1, 2, \ldots, n$. $\endgroup$ – Adayah Jul 31 '17 at 14:32
  • $\begingroup$ Your $k$'s turned into $n$'s at the end, $\endgroup$ – zhw. Jul 31 '17 at 14:34
  • $\begingroup$ @Adayah Okay, thanks! $\endgroup$ – Franklin Pezzuti Dyer Jul 31 '17 at 14:35
  • $\begingroup$ @zhw Whoops, thank you! $\endgroup$ – Franklin Pezzuti Dyer Jul 31 '17 at 14:36

The other users basically already answered your question, I just want to recommend that you try showing this by proving that the preimage of any open set is open. This would save you a lot of time. You could also prove this by taking the limits and showing they exist. Both of these are much shorter.


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.