50
$\begingroup$

What is the "standard" way to denote all positive (or non-negative) real numbers? I'd think

$$ \mathbb R^+ $$

but I believe that that is usually used to denote "all real numbers including infinity".

So is there a standard way to denote the set

$$ \{x \in \mathbb R : x \geq 0\} \; ?$$

$\endgroup$
  • 16
    $\begingroup$ Note that $0$ is not positive. $\endgroup$ – Yuval Filmus Mar 19 '11 at 15:08
  • 23
    $\begingroup$ Also, I wouldn't agree that $R_+$ usually includes $\infty$. The extended real line is used only in certain areas. $\endgroup$ – Yuval Filmus Mar 19 '11 at 15:09
  • 2
    $\begingroup$ I removed the set theory tag since this isn't a set theory question. $\endgroup$ – Apostolos Mar 19 '11 at 15:09
  • 5
    $\begingroup$ $[0,\infty)$ or if you want to work with the extended real line, $[0, +\infty]$. $\endgroup$ – cardinal Mar 19 '11 at 15:12
  • 5
    $\begingroup$ @YuvalFilmus Do not forget that this is just an english convention. In France for example, we usually say that 0 is both positive and negative. I have often seen $\mathbb{R}^+$ for all positive/null numbers and $\mathbb{R}^{\ast +}$ for all strictly positive numbers. $\endgroup$ – ThR37 Jun 17 '14 at 9:53
33
$\begingroup$

Not that I knew of. There are many, e.g.

  • $\mathbb{R^+_0}$,
  • $\mathbb{R^+}$ and
  • $[0, \infty)$.
$\endgroup$
34
$\begingroup$

The unambiguous notations are: for the positive-real numbers $$ \mathbb{R}_{>0} = \left\{ x \in \mathbb{R} \mid x > 0 \right\} \;, $$ and for the non-negative-real numbers $$ \mathbb{R}_{\geq 0} = \left\{ x \in \mathbb{R} \mid x \geq 0 \right\} \;. $$ Notations such as $\mathbb{R}_{+}$ or $\mathbb{R}^{+}$ are non-standard and should be avoided, becuase it is not clear whether zero is included. Furthermore, the subscripted version has the advantage, that $n$-dimensional spaces can be properly expressed. For example, $\mathbb{R}_{>0}^{3}$ denotes the positive-real three-space, which would read $\mathbb{R}^{+,3}$ in non-standard notation.


Addendum:

In Algebra one may come across the symbol $\mathbb{R}^\ast$, which refers to the multiplicative units of the field $\big( \mathbb{R}, +, \cdot \big)$. Since all real numbers except $0$ are multiplicative units, we have $$ \mathbb{R}^\ast = \mathbb{R}_{\neq 0} = \left\{ x \in \mathbb{R} \mid x \neq 0 \right\} \;. $$ But caution! The positive-real numbers can also form a field, $\big( \mathbb{R}_{>0}, \cdot, \star \big)$, with the operation $x \star y = \mathrm{e}^{ \ln(x) \cdot \ln(y) }$ for all $x,y \in \mathbb{R}_{>0}$. Here, all positive-real numbers except $1$ are the "multiplicative" units, and thus $$ \mathbb{R}_{>0}^\ast = \left\{ x \in \mathbb{R}_{>0} \mid x \neq 1 \right\} \;. $$

$\endgroup$
  • 6
    $\begingroup$ The last objection makes no sense since one could simply use $\mathbb R_+^3$. $\endgroup$ – Did Oct 25 '15 at 10:51
  • $\begingroup$ Actually for $\mathbb R^+\times\mathbb R^+\times\mathbb R^+$ I'd write $(\mathbb R^+)^3$. The notation with comma doesn't look right to me. $\endgroup$ – celtschk Jun 13 '17 at 6:08
  • 1
    $\begingroup$ Would that itwere so simple. In Probability with Martingales Williams tells me "Everyone is agreed that $\mathbb{R}^+$ is $[0,\infty)$. $\endgroup$ – Addem Oct 17 '17 at 19:38
9
$\begingroup$

I'd completely avoid using $\mathbb{R}^+$ since people won't know if $0$ is included or not. So $\mathbb{R}_0^+$ would be a possibility, but then how would you denote $\{x\in\mathbb{R}:x>0\}$? Again, with $\mathbb{R}^+$ people won't know that $0$ isn't included. Personally, I prefer writing $[0,\infty)$ and $(0,\infty)$ when it's clear from the context that an interval in $\mathbb{R}$ is meant.

$\endgroup$
  • 1
    $\begingroup$ All the mathematicians I ever met , ( a lot), understood that $R^+$ meant the positive reals. $\endgroup$ – DanielWainfleet Aug 25 '15 at 20:20
  • 9
    $\begingroup$ @user254665: Well, certainly it means the positive reals, but now ask them what they mean by "positive" :-) Seriously, I know mathematicians who mean "$\ge0$" and other who mean "$>0$". $\endgroup$ – Hendrik Vogt Aug 28 '15 at 17:26
  • $\begingroup$ Edit: I think that $\mathbb{R}^{+} \backslash \Bigl\{\left((\mathbb{R}^{+} \backslash \mathbb{R}_0^{+}) \cup (\mathbb{R}_0^{+} \backslash \mathbb{R}^{+})\right)\Bigr\} \cup \{1\}$ will be unambiguous :-) $\endgroup$ – Kusavil Jan 28 '18 at 1:13
6
$\begingroup$

Some of my profs use $\mathbb{R^{\ge 0}}$. I like to add whatever to the top so $\mathbb{R^{\le a}}$ just means all reals less than $a$.

$\endgroup$
  • 7
    $\begingroup$ This definitely strikes me as nonstandard, at least in the U.S. I'd be curious to know where all this is used. (Not saying it's a bad notation, just never seen it in any texts of common mathematics publishers, for example.) $\endgroup$ – cardinal Mar 19 '11 at 18:55
  • 1
    $\begingroup$ I learned this from my math prof who grew up in Canada. But yeah I've never seen it outside her notes, but it does make writing $\{ x \in R \mid x < a\}$ easier! $\endgroup$ – hwong557 Mar 19 '11 at 19:00
  • 1
    $\begingroup$ Interval notation does not per se fix the basic set. $\endgroup$ – Raphael Mar 19 '11 at 21:11
  • 1
    $\begingroup$ @cardinal: I've seen it used many times in Europe (but rather as subscript: $\mathbb{R}_{\geq 0}$) and some people even write $\mathbb{Z}_{\geq 0}$ instead of $\mathbb{N}$ because the latter is ambiguous as to whether $0$ is in it or not. And of course all obvious variants such as $\mathbb{R}_{\lt t}$ and so on are also used. But certainly, interval notation is more common. $\endgroup$ – t.b. Mar 21 '11 at 3:00
  • 2
    $\begingroup$ @cardinal: I think I can confirm that to a certain extent. I'm pretty sure we exclusively used interval notation à la Bourbaki in elementary and high school in Switzerland (I had at least 6 math teachers at various places) and it is exclusively used in at least four elementary texts on (what we call) algebra in my bookshelf. $\endgroup$ – t.b. Mar 21 '11 at 13:30
4
$\begingroup$

The following is also pretty common notation for the non-negative reals: $\mathbb{R}_{\geq 0}$ or $\mathbb{R}_{+}$.

$\endgroup$
2
$\begingroup$

I've learned in elementary school that $\mathbb{R}_{*}$ means the set without the zero, so $\mathbb{R}^{+}=[0,\infty)$ and $\mathbb{R}^{+}_{*}=(0,\infty)$.

$\endgroup$
  • $\begingroup$ And I learned in school that $\mathbb R^+ = (0,\infty)$, and $R_0^+ = [0,\infty)$. Well, except that we would have written those intervals as $]0;\infty[$ and $[0;\infty[$ … $\endgroup$ – celtschk Jun 13 '17 at 6:21
0
$\begingroup$

$\mathbb{R}^+$ includes $0$ in Probability Tutorials. $\mathbb{R}^+_0$ is more clear though, so I've used it in the exercises.

$\endgroup$
0
$\begingroup$

I find $\mathbb R_{\geq 0}$ clumsy (I would never write this on a board when working and I don't often see papers writing functions $f$ defines as $f:\mathbb R_{\geq 0}\rightarrow \mathbb R_{\geq 0}$).

$\mathbb R^+$ seems restrictive, not least if you wish to consider higher dimensions.

I like $[0,\infty)$, but it can be awkward in certain settings such as $f:[0,\infty)\times (0,\infty)\rightarrow \mathbb [0,\infty)$ or

$$\left\{E\times[0,\infty)\times (0,\infty)\right\}$$

Instead I prefer $\mathbb{\bar R_+}$ for the nonnegative reals and $\mathbb R_+$ for the positive reals. This fits with the notion of closure in $\mathbb R$. (This might not suit those who regularly deal with the extended reals, but given that $\mathbb R$ is so standard, it seems natural to take the closure there.) The function $f: \mathbb{\bar R_+}\rightarrow \mathbb{\bar R_+}$ is then clear and reasonably compact. Moreover, $$\left\{E\times\mathbb{\bar R_+}\times \mathbb R_+\right\}$$ and $f: \mathbb{\bar R_+}\times \mathbb R_+\rightarrow \mathbb{\bar R_+}$ seem to be substantially easier to read than the interval versions above.

Consistency then dictates that $\mathbb Z_+$ denotes the positive integers and whilst $\mathbb {\bar Z_+}$ is arguably unsatisfactory notation for the nonnegative integers because the closure story no longer applies, I would adopt it in order to be consistent. You could use $\mathbb N=\mathbb Z_+\cup\{0\}$, but that seems worse.

I guess it depends on the problem at hand.

ps. I have also seen $\mathbb R_{++}$ for the positive reals and $\mathbb R_+$ for the nonnegative.

$\endgroup$

protected by Zev Chonoles Sep 2 '16 at 2:38

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.