1

Let $f(x)= (x² + 1) / (√x²-1)$. Determine the domain of f and prove that f is continuous for every point in its domain.

So domain would be determined by what makes $(√x²−1) ≠ 0$? This would be all of $x∈R$ except $x≠1$ or $x≠ -1$. Then you would have to prove continuity for the intervals $(-∞,-1), (-1, 1),$ and $(1, ∞)$. And $(x²+1)>0$ simply by its nature. And $(√x²-1)>0$ again because a square root cannot be negative. Thus the domain of $f$ is $(1, ∞)$.

Proof has to be made using Epsilon-Delta definition: $∀ε>0 ∃δ>0$ such that if $x∈I$ and $|x−x₀|<δ$ then $|f(x)−f(x₀)|<ε$.

Let $g(x)=x²+1$. And let $h(x)=√(x²-1)$. If we can show that each of these is continuous for all $x\in (1,∞)$, then by the rules of continuity, we can show $f=g/h$ is continuous as well.

Let's start with $g$.

Proof: Let $g(x)=x²+1$.

Let $\epsilon>0$ be arbitrary. Let $x₀\in (1, ∞)$ be arbitrary.

Let $\delta g=min ❴1, \epsilon/(2|x₀|+1)❵$.

Let $x\in (1, ∞)$ and $|x-x₀|<\delta g$.

Thus $|g(x)-g(x₀)|=|x²+1-x₀²-1|=|x²-x₀²|<(2|x₀|+1)|x-x₀|<(2|x₀|+1)\delta g\leq(2|x₀|+1)(\epsilon/(2|x₀+1))=\epsilon$

Thus $|g(x)-g(x₀)|<\epsilon$. Therefore since $\epsilon$ is arbitrary, $g$ is continuous at $x₀$. And since $x₀$ is arbitrary, $g$ is continuous for all of $x\in (1, ∞)$. ∎

Now $h$.

Proof: Let $h(x)=√(x²-1)$.

Let $\epsilon>0$ be arbitrary. Let $x₀\in (1, ∞)$ be arbitrary.

Let $\delta h=?$.

Let $x\in (1, ∞)$ and $|x-x₀|<\delta h$.

Thus $|h(x)-h(x₀)|=|√(x²-1)-√(x₀²-1)|=?$ And this is where I get stuck.

The rest would go:

Thus $|h(x)-h(x₀)|<\epsilon$. Therefore since $\epsilon$ is arbitrary, $h$ is continuous at $x₀$. And since $x₀$ is arbitrary, $h$ is continuous for all of $x\in (1, ∞)$. ∎

Let $f(x)=g(x)/h(x)$.

Because $h(x)≠0$ for $x\in (1, ∞)$ then f is continuous for all $x\in (1, ∞)$∎

Maddy
  • 193
  • Welcome to math.SE! Here's a quick guide on how to write math on this site: http://meta.math.stackexchange.com/questions/5020/mathjax-basic-tutorial-and-quick-reference – Newb Oct 29 '13 at 16:38
  • Could someone please help me with this one? – Maddy Nov 03 '13 at 09:19
  • @Newb: What exactly is wrong with the way it is written? – Maddy Nov 03 '13 at 09:20
  • Nothing now - you're correct about the domain. Note also that you cannot have the square root (in $\mathbb{R}$) of a negative number, so you'll have to adjust appropriately. Have you done any work with the $\epsilon-\delta$ proof? – Newb Nov 03 '13 at 14:53
  • @Newb Thank you for that. I missed that totally. So the domain would be $(0,1)$ and $(1,∞)$ all of it being non-inclusive of $0$ and $1$. – Maddy Nov 03 '13 at 19:09
  • @Newb Made an attempt on the proof. – Maddy Nov 05 '13 at 02:42
  • I don't actually think I'm qualified to tell you whether your proof works or not - I don't yet know how to prove continuity of functions. Sorry! – Newb Nov 05 '13 at 05:12

0 Answers0