Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Given $A = \{(x,y) \in \mathbb{R^2}: y = \frac{1}{x}, x > 0\}$. Show (by considering convergent sequences or otherwise) $A$ is closed in $\mathbb{R^2}$.

Anyone able to give me some advice on how to approach this problem. I know that if every convergent sequence $(x_n, y_n) \in A$ converges in $A$ then $A$ is closed. But I don't know what process I should use to show this and I don't know how to deal with sequences that use two terms, ie $x_n, y_n$ instead of just one term.

share|cite|improve this question
up vote 3 down vote accepted

HINT: Suppose that $\langle p_n:n\in\Bbb N\rangle$ is a sequence in $A$ that converges in $\Bbb R^2$. Then there are sequences $\langle x_n:n\in\Bbb N\rangle$ and $\langle y_n:n\in\Bbb N\rangle$ in $\Bbb R$ such that $p_n=\langle x_n,y_n\rangle$ for each $n\in\Bbb N$. Let $p=\langle x,y\rangle\in\Bbb R^2$ be the limit of the sequence $\langle p_n:n\in\Bbb N\rangle$; you want to show that $p\in A$.

For each $n\in\Bbb N$, $p_n=\langle x_n,y_n\rangle\in A$, so we know that $x_n>0$ and $y_n=\dfrac1{x_n}$, so that $$p_n=\left\langle x_n,\frac1{x_n}\right\rangle\;.$$

  • Use the fact that $\langle p_n:n\in\Bbb N\rangle\to\langle x,y\rangle$ to show that $\langle x_n:n\in\Bbb N\rangle\to x$ and $\langle y_n:n\in\Bbb N\rangle\to y$.

  • Use the continuity of the function $f(x)=\frac1x$ to conclude that $y=\frac1x$ and hence that $\langle x,y\rangle\in A$.

An alternative approach is to show that $\Bbb R^2\setminus A$ is open: show that if $p=\langle x,y\rangle\notin A$, then there is an $r>0$ such that $B(p,\epsilon)\cap A=\varnothing$, where $B(p,\epsilon)$ is the open ball of radius $\epsilon$ centred at $p$. Note that any $\epsilon$ less than or equal to the distance from $p$ to $A$ will work.

share|cite|improve this answer
As $p = (x,y)$ is the limit of $p_n = (x_n,y_n)$ can we just say it is clearly true that $x_n \to x$ and $y_n \to y$ as $p$ is composed of those two sequences and $p_n \to (x,y)$? I'm not sure why we need to use the continuity of $f(x)$ to show that $y=\frac{1}{x}$, aren't we given that fact in the question itself? The open ball approach also seems like a good idea but how do you take the distance from a point to a set? – sonicboom Nov 7 '12 at 0:34
@sonicboom: ‘Clearly true’? No, it has to be proved. One way to do so is to let $\pi_1:\Bbb R^2\to\Bbb R:\langle x,y\rangle\mapsto x$, the projection to the $x$-axis; prove that $\pi_1$ is continuous, and note that $\langle x_n:n\in\Bbb N\rangle$ is just $\langle\pi_1(p_n):n\in\Bbb N\rangle$, while $x=\pi_1(p)$. Then do something similar for the other coordinate. You’re certainly not given that $y=\frac1x$: that is in fact precisely what you need to prove! This is a specific $\langle x,y\rangle$, the limit of the $p_n$’s, and you want to prove that it’s in $A$, which means proving that $xy=1$. – Brian M. Scott Nov 7 '12 at 0:40
@sonicboom: The sequence approach is probably a bit easier than computing the distance from an arbitrary point to that hyperbola. – Brian M. Scott Nov 7 '12 at 0:41
I need to think this one over in depth. Cheers for the pointers. – sonicboom Nov 7 '12 at 1:09
I think I have a better understanding of this one now...I get what is happening with using the continuity of $f(x) = \frac{1}{x}$ to show $y = \frac{1}{x}$ by continuity of $f$, $(x_n) \to x$ implies $f(x_n) \to f(x)$, which is $y_n \to \frac{1}{x}$ and we already have $(y_n) \to y$ so $y=\frac{1}{x}$. I am 90% there on why we need to show $(x_n) \to x$ and $(y_n) \to y$ as opposed to it being clearly true, need to think that over a bit more. – sonicboom Nov 7 '12 at 11:50

Note that $A$ consists of all $(x,y)$ such that $xy-1=0$, so it is the preimage of $0$ under the continuous function $f(x,y)=xy-1$.

share|cite|improve this answer
...and since $\{0\}$ is closed in $\mathbb{R}$, the preimage of $\{0\}$ must be closed in $\mathbb{R}^2.$ – Fly by Night Nov 6 '12 at 21:15
...actually, to get $A$ you also have to intersect the preimage with the (closed) set $\{(x,y)\mid x\ge 0\}$. – nonpop Nov 6 '12 at 21:22

I would start by simplifying the problem. Your graph is homeomorphic to the positive real numbers. Take the map $f : \mathbb{R}^2 \to \mathbb{R}^2$ given by $f : (x,y) \mapsto (x,y-\frac{1}{x}).$ This map is in fact a diffeomorphism for all $x \neq 0.$ (A diffeomorphism is a differentiable homeomorphism.)

This map was chosen because $ f : (t,\frac{1}{t}) \mapsto (t,\frac{1}{t}-\frac{1}{t}) = (t,0).$ Thus, your graph is sent to the positive real axis. Now you need only show that this is closed in $\mathbb{R}^2.$ Well, you should easily be able to show (using open discs) that the complement of this set is open, and so it itself if closed.

share|cite|improve this answer

Your Answer


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.