0
$\begingroup$

Let $S = [a,b]$ where $0 < a \leq b$ and $a,b \in \mathbb{N}$. Meaning all possible sets where this is true (e.g., $[1,1],[1,2],[5,5],[5,n],\cdots$)

Alright, this isn't for an assignment, just a question I arrived at and have been having difficulty figuring out. My original thought was that this must have the same cardinality as $\mathbb{Q}$, but then when attempting to find a bijection between the two sets , I ran into an issue.

My first attempt,

for all $s \in S$ let $s = a/b$ (first ,last number in the set be assigned to that rational). The problem with this is that I cannot arrive at any rational bigger than $1$. If I reverse the order ( $s = b/a$), I cannot get to any number smaller than $1$.

My second attempt was to try and find some way to use the diagonalization argument , but again, if I block the numbers in either way as I stated above, I cannot reach a large set of numbers (either smaller than or bigger than zero).

So since this set extends $\mathbb{N}$ I assume its cardinality must be at least that. But I cannot find a bijection between $\mathbb{N}$ and $S$ or between $\mathbb{Q}$ and $S$.

EDIT, I don't think that i was clear that this was a set of sets. So things so the set $S$ would contain :

$$[1,1],[1,2],[1,3],\cdots [2,2],[2,3],[2,3],\cdots$$

$\endgroup$
2
  • $\begingroup$ Is $S$ a set of consecutive integers, or an interval of real numbers? $\endgroup$ – Arthur Sep 16 '16 at 17:16
  • $\begingroup$ S is a set of sets. All sets [a,b] where a,b are integers and 0<a<=b $\endgroup$ – razeal113 Sep 18 '16 at 21:13
4
$\begingroup$

If $a=b$, $S=\{a\}$, so its cardinality is $1$

If $a<b$, its cardinality is the same as the cardinality of $[0,1]$ since there is a linear function $f$ such that $f(a)=0$, $f(b)=1$. Now $\DeclareMathOperator\card{card}\card\,(0,1)=\card\mathbf R$, hence the $\card\,[0,1]\ge\card\mathbf R$. But as it is contained in $\mathbf R$, we also have $\card\,[0,1]\le\card\mathbf R$, whence $$\card\,[0,1]=\card\mathbf R.$$

Added: Answer to the updated question – the cardinality of $\;S=\{[a,b]\mid 0 <a\le b,\;a, b\in\mathbf N \}$. There's a injection $\begin{aligned}[t]S&\longrightarrow\mathbf N\times\mathbf N\\{}[a,b]&\longmapsto (a,b)\end{aligned},\;$ which is countable. Hence $S$ is at most countable. As it is infinite, it is countable.

$\endgroup$
6
  • $\begingroup$ +1. Note for the OP: we can see that $(0, 1)$ and $\mathbb{R}$ have the same cardinality by given an explicit bijection between them - for example, consider $$f(x)={(\arctan(x)+{\pi\over 2})\over \pi}.$$ Also note for the OP, this means that we can have $A$ and $B$ have the same cardinality even when $A$ is a proper subset of $B$! $\endgroup$ – Noah Schweber Sep 16 '16 at 18:07
  • $\begingroup$ @Noah Schweber; Maybe $f(x)=\dfrac{\mathrm e^x}{1+\mathrm e^x}$ is more intuitive? Or $\dfrac{1+x+\lvert x\rvert}{2(1+\lvert x\rvert)}$. $\endgroup$ – Bernard Sep 16 '16 at 18:46
  • $\begingroup$ Those also work, but I think the one I presented is clearer, since it is more transparent (IMHO) where the idea comes from: take a function you already know bijects $\mathbb{R}$ with an interval, then shift and scale appropriately. $\endgroup$ – Noah Schweber Sep 16 '16 at 18:52
  • $\begingroup$ That's what I did with the last one, as $\dfrac x{1+\lvert x\rvert}$ is a bijection onto $(-1,1)$ well-known to many people. Many beginners are reluctant with trigonometric functions, I think – but maybe I'm wrong. $\endgroup$ – Bernard Sep 16 '16 at 18:57
  • $\begingroup$ Sorry I don't think i was clear enough about this question; i added an edit to hopefully make the question more clear $\endgroup$ – razeal113 Sep 18 '16 at 11:14

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.